Code for the manuscript "Machine learning reveals STAT motifs as predictors for GR-mediated gene repression"
Help improve this workflow!
This workflow has been published but could be further improved with some additional meta data:- Keyword(s) in categories input, output, operation, topic
You can help improve this workflow by suggesting the addition or removal of keywords, suggest changes and report issues, or request to become a maintainer of the Workflow .
Code for the manuscript "Machine learning reveals STAT motifs as predictors for GR-mediated gene repression"
TODO: Add citation once paper is accepted
Table of contents
Code Snippets
54 55 | shell: "wget {params.url} -P {params.outdir}" |
68 69 | shell: "bowtie-build -f {input[fa]} {params.outdir}/{params.basename}" |
81 82 83 84 85 86 87 88 89 90 91 92 | shell: """ wget http://mitra.stanford.edu/kundaje/akundaje/release/blacklists/mm10-mouse/mm10.blacklist.bed.gz -P {params.outdir} && \ wget http://ftp.ensembl.org/pub/release-100/gtf/mus_musculus/Mus_musculus.GRCm38.100.gtf.gz -P {params.outdir} && \ wget http://hgdownload.cse.ucsc.edu/goldenPath/mm9/liftOver/mm9ToMm10.over.chain.gz -P {params.outdir} && \ gunzip data/current/mm9ToMm10.over.chain.gz && \ mkdir -p "data/current/gencode_annotations" && \ wget https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_mouse/release_M1/gencode.vM1.annotation.gtf.gz -P {params.outdir}/gencode_annotations/ && \ wget https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_mouse/release_M23/gencode.vM23.annotation.gtf.gz -P {params.outdir}/gencode_annotations/ && \ zcat data/current/gencode_annotations/gencode.vM23.annotation.gtf.gz | grep -P '\tgene\t' > data/current/gencode_annotations/gencode.vM23.annotation.gene.gtf """ |
104 105 106 107 108 109 110 111 112 | shell: """ mkdir -p {params.outdir} && \ wget {params.downloadlink} -P {params.outdir} && \ gunzip -c {params.outdir}/remap2022_all_macs2_mm10_v1_0.bed.gz > {output.fullbed} && \ sed -E "s/(\\w+)\\.(\\w+)\\.(\\w+)/\\1\\t\\2\\t\\3/g" {output.fullbed} > {output.separatedfulltsv} && \ awk -F'\t' '$6~/BMDM|macrophage/' {output.separatedfulltsv} > {output.filteredtsv} && \ awk '{{print $1"\t"$2"\t"$3"\t"$4","$5","$6}}' FS='\t' {output.filteredtsv} > {output.filteredbed} """ |
119 120 121 122 123 124 125 | shell: "gunzip -c {input} > {output}" # -c keeps the original file unchanged rule scan_nr3c1_genomewide: input: motif="data/current/motifs/custom/nr3c1_simplified_{sitelength}.motif", fa= "data/current/genome/GRCm38.primary_assembly.standchr.fa" |
128 129 130 131 132 | shell: """ mkdir -p $(dirname {output}) && \ scanMotifGenomeWide.pl {input.motif} {input.fa} -bed > {output} """ |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--ABC_DexLPS_all"), type="character", help="Path to file with all ABC predictions passing 0.02 (in DexLPS condition)"), make_option(c("--ABC_LPS_all"), type="character", help="Path to file with all ABC predictions passing 0.02 (in LPS condition)"), make_option(c("--fimo_results_summitregion"), type="character", help="Path to rds file of fimo motifcounts within summitregions"), make_option(c("--fimo_results_dexlps"), type="character", help="Path to rds file of fimo motifcounts within ABC regions (in DexLPS condition)"), make_option(c("--fimo_results_lps"), type="character", help="Path to rds file of fimo motifcounts within ABC regions (in LPS condition)"), make_option(c( "--chipseq_ranges"), type="character", help="Path to summit file of IDR peaks"), make_option(c("--outdir"), type="character", help="Path to output directory") ) opt <- parse_args(OptionParser(option_list=option_list)) dir.create(opt$outdir) #change default for stringAsFactors options(stringsAsFactors = FALSE) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=10, family = "ArialMT", colour="black"), title=element_text(size=10, family="ArialMT", colour="black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text = element_text(size=10, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black")) set.seed(12345) #------------------------------- ## read in data #------------------------------- ABC_DexLPS_all <- read.delim(opt$ABC_DexLPS_all) %>% plyranges::as_granges(., seqnames=chr) ABC_LPS_all <- read.delim(opt$ABC_LPS_all) %>% plyranges::as_granges(., seqnames=chr) fimo_results_dexlps <- readRDS(opt$fimo_results_dexlps) fimo_results_lps <- readRDS(opt$fimo_results_lps) fimo_results_summitregion <- readRDS(opt$fimo_results_summitregion) chipseq_ranges <- readRDS(opt$chipseq_ranges) # assign a name to allow for matching for prox based assignment later chipseq_ranges$name <- paste(seqnames(chipseq_ranges), start(chipseq_ranges), end(chipseq_ranges), sep="_") #-------------------------------------------- #-------------------------------------------- ## prepare motifcounts #-------------------------------------------- #-------------------------------------------- ## summitregions #-------------------------------------------- get_summitregion_motifcounts <- function(summits, fimo_results){ # Take 100bp windows around ChIP-seq summits to recreate the original query fimo_queries_summitregion <- summits %>% plyranges::anchor_center() %>% plyranges::mutate(width = 100) ## intersect queries with fimo hits summitregion_leftjoin_query_hits <- fimo_queries_summitregion %>% plyranges::join_overlap_left(fimo_results) # aggregate motifcounts per query # seqnames is needed for the train test split later summitregion_leftjoin_query_hits_motifsaggregated <- as.data.frame(summitregion_leftjoin_query_hits) %>% group_by(name, seqnames, motif_alt_id) %>% summarize(motifcount = n()) # if a region has no motifmatches at all, it get's an NA which showes up as motifname after doing pivor_wider # cast it in a way, so we have unique regions as rows and all observed motifs as columns motifcounts <- summitregion_leftjoin_query_hits_motifsaggregated %>% tidyr::pivot_wider(names_from = motif_alt_id, values_from = motifcount, values_fill = 0) %>% dplyr::select(!'NA') # remove NA motif that got introduced by region without any matches return(motifcounts) } motifcounts_summitregion <- get_summitregion_motifcounts( chipseq_ranges, fimo_results_summitregion ) ## ABC enhancerregions #-------------------------------------------- get_ABC_motifcounts <- function(ABC_results, fimo_results){ # intersect queries with fimo hits ABC_results_unique <- ABC_results %>% unique() ABC_results_leftjoin_query_hits <- ABC_results_unique %>% plyranges::join_overlap_left(fimo_results) # aggregate motifcounts per query ABC_results_leftjoin_query_hits_motifsaggregated <- as.data.frame(ABC_results_leftjoin_query_hits) %>% group_by(name, seqnames ,motif_alt_id) %>% summarize(motifcount = n()) # cast it in a way, so we have unique regions as rows and all observed motifs as columns motifcounts_abcregion <- ABC_results_leftjoin_query_hits_motifsaggregated %>% tidyr::pivot_wider(names_from = motif_alt_id, values_from = motifcount, values_fill = 0) %>% dplyr::select(!'NA') # remove NA motif that got introduced by region without any matches return(motifcounts_abcregion) } motifcounts_abcregion_dexlps <- get_ABC_motifcounts( ABC_DexLPS_all, fimo_results_dexlps ) motifcounts_abcregion_lps <- get_ABC_motifcounts( ABC_LPS_all, fimo_results_lps ) #-------------------------------------------- ## ASSIGNMENTS #-------------------------------------------- ## prox based #-------------------------------------------- # these are mgi_symbols, the ones from ABC are ensemble IDs. Is this a problem? assignment_summit_prox <- as.data.frame(chipseq_ranges) %>% dplyr::rename(anno=mgi_symbol) %>% dplyr::select(name, anno) ## hybrid #-------------------------------------------- # we need an assignment of the regionIDs (from the ChIPseq summtiregion) # to the ABC derived assignments get_assignment_summit_ABCregion <- function(summit, ABCregions){ # what summit lies in what ABC region assignment_summit_ABCregion <- summit %>% plyranges::join_overlap_left(ABCregions) assignment_summit_ABCregion <- as.data.frame(assignment_summit_ABCregion) %>% dplyr::select(name.x, TargetGene, ABC.Score, ABC.Score.Numerator, class, isSelfPromoter) %>% magrittr::set_colnames(c("name","anno","abcscore","abcnumerator","class","isSelfPromoter")) return(assignment_summit_ABCregion) } assignment_summits_ABCregion_dexlps <- get_assignment_summit_ABCregion( chipseq_ranges, ABC_DexLPS_all ) assignment_summits_ABCregion_lps <- get_assignment_summit_ABCregion( chipseq_ranges, ABC_LPS_all ) ## ABC based #-------------------------------------------- assignment_abcregion_dexlps <- as.data.frame(ABC_DexLPS_all) %>% dplyr::select(name,TargetGene,ABC.Score,ABC.Score.Numerator,class,isSelfPromoter) %>% magrittr::set_colnames(c("name","anno","abcscore","abcnumerator","class","isSelfPromoter")) %>% dplyr::mutate(anno=gsub("\\.[0-9]*$","",anno)) assignment_abcregion_lps <- as.data.frame(ABC_LPS_all) %>% dplyr::select(name,TargetGene,ABC.Score,ABC.Score.Numerator,class,isSelfPromoter) %>% magrittr::set_colnames(c("name","anno","abcscore","abcnumerator","class","isSelfPromoter")) %>% dplyr::mutate(anno=gsub("\\.[0-9]*$","",anno)) #-------------------------------------------- ## export objects #-------------------------------------------- # assignments saveRDS(assignment_summit_prox, paste0(opt$outdir,"assignment_summit_prox.rds")) saveRDS(assignment_summits_ABCregion_dexlps, paste0(opt$outdir,"assignment_summits_ABCregion_dexlps.rds")) saveRDS(assignment_summits_ABCregion_lps, paste0(opt$outdir,"assignment_summits_ABCregion_lps.rds")) saveRDS(assignment_abcregion_dexlps, paste0(opt$outdir,"assignment_abcregion_dexlps.rds")) saveRDS(assignment_abcregion_lps, paste0(opt$outdir,"assignment_abcregion_lps.rds")) # motifcoutns saveRDS(motifcounts_summitregion, paste0(opt$outdir,"motifcounts_summitregion.rds")) saveRDS(motifcounts_abcregion_dexlps, paste0(opt$outdir,"motifcounts_abcregion_dexlps.rds")) saveRDS(motifcounts_abcregion_lps, paste0(opt$outdir,"motifcounts_abcregion_lps.rds")) |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--contrast_DexLPSvLPS"), type="character", help="Path to annotated tsv file of DeSeq2 contrast of DexLPS vs LPS"), make_option(c("--assignment_summit_prox"), type="character", help="Path to rds file of proximity based assignment of peak summits to genes"), make_option(c("--assignment_summits_abcregion_dexlps"), type="character", help="Path to rds file of assignment of peak summits within abcregions to genes (in DexLPS condition)"), make_option(c("--assignment_summits_abcregion_lps"), type="character", help="Path to rds file of assignment of peak summits within abcregions to genes (in LPS condition)"), make_option(c("--assignment_abcregion_dexlps"), type="character", help="Path to rds file of assignment of abcregions to genes (in DexLPS condition)"), make_option(c( "--assignment_abcregion_lps"), type="character", help="Path to rds file of assignment of abcregions to genes (in LPS condition)"), make_option(c("--motifcounts_summitregion"), type="character", help="Path to rds file of fimo motifcounts within summitregions"), make_option(c("--motifcounts_abcregion_dexlps"), type="character", help="Path to rds file of fimo motifcounts within ABC regions (in DexLPS condition)"), make_option(c("--motifcounts_abcregion_lps"), type="character", help="Path to rds file of fimo motifcounts within ABC regions (in LPS condition)"), make_option(c( "--outdir"), type="character", help="Path to output directory") ) opt <- parse_args(OptionParser(option_list=option_list)) dir.create(opt$outdir) #change default for stringAsFactors options(stringsAsFactors = FALSE) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(org.Mm.eg.db, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(DESeq2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ComplexHeatmap, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=6, family = "ArialMT", colour="black"), title=element_text(size=8, family="ArialMT", colour="black"), panel.grid.major = element_line(colour="grey", size=0.2), panel.grid.minor = element_blank(), axis.text = element_text(size=6, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(6, 'points'), #change legend key size legend.key.height = unit(6, 'points'), #change legend key height legend.key.width = unit(6, 'points'), #change legend key width legend.text = element_text(size=6, family="ArialMT", colour="black")) set.seed(12345) #------------------------------- ## read in data #------------------------------- contrast_DexLPSvLPS <- read.delim(opt$contrast_DexLPSvLPS) for (optname in names(opt)[2:9]){ #except for the outdir print(paste0("Loading ", optname)) assign(optname, readRDS( opt[[optname]] )) } #-------------------------------------------- ## ---- function definitions #-------------------------------------------- coerce_coef2df <- function(model_coef){ model_coef <- as.matrix(model_coef) model_coef <- as.data.frame(model_coef) model_coef$names <- rownames(model_coef) colnames(model_coef) <- c("estimates", "names") return(model_coef) } #------------function will------------: # * merge the motifdata with gene assignments and # * then merge the gene expression changes, # * filter for DE genes and aggregate per gene #-------------------------------------- merge_motifdata_with_assignments <- function( motifcounts, assignments, contrast, maxonly=FALSE, excludepromoters=FALSE, weightby=FALSE, sepPromEnh=FALSE){ # check if we should use all abc assignments, or only the max one of each peakID if(maxonly==TRUE){ assignments <- assignments %>% group_by(name) %>% filter(abcscore==max(abcscore)) %>% distinct() } else{ assignments <- assignments } if(excludepromoters=="all"){ assignments <- assignments %>% filter(!class=="promoter") } else if (excludepromoters=="onlyNONself"){ assignments <- assignments %>% filter(!c(class=="promoter" & isSelfPromoter=="False")) } else{ assignments <- assignments } motifdf <- merge(motifcounts, assignments, by.x="name", by.y="name") motifdf <- motifdf %>% relocate(c(anno)) # merge the expression change # use mgi_symbol for prox based, otherwise ensemblID # We DONT set all.x=TRUE because we don't care about predicting gene that aren't even expressed or that we don't have a clear label for if(maxonly=="prox"){ motifdf <- merge(motifdf, contrast, by.x="anno", by.y="mgi_symbol") } else { motifdf <- merge(motifdf, contrast, by.x="anno", by.y="Row.names") } #recode the logFC and padj into a label (optionally through command line arguments) motifdf <- motifdf %>% mutate(label=case_when(log2FoldChange>0.58 & padj < 0.05 ~ "up", log2FoldChange<(-0.58) & padj < 0.05 ~ "down", TRUE ~ "no_change")) %>% filter(label!="no_change") %>% mutate(label=factor(label, levels=c("down","up"), labels=c(0,1))) %>% relocate(label) # chr 2, 3 and 4 (20%) were used as the tuning set for hyperparameter tuning. # Regions from chromosomes 1, 8 and 9 (20%) were used as the test set for performance evaluation # The remaining regions were used for model training. #------aggregate over gene SYMBOL if ("abcscore" %in% colnames(motifdf)) { # loop for ABC based assignments if (weightby=="abcscore"){ unselect_col <- "abcnumerator" } else if(weightby=="abcnumerator") { unselect_col <- "abcscore" } else{ unselect_col <- c("abcscore","abcnumerator") } if (sepPromEnh==TRUE){ motifdf_aggr <- motifdf %>% dplyr::select(!c(unselect_col,"name","baseMean","log2FoldChange","lfcSE","stat","pvalue","padj","gene_biotype","mgi_symbol")) %>% { if(weightby!=FALSE) mutate(.,across(where(is.numeric), ~ (.x * get(weightby)))) else .} %>% # weight features by score dplyr::select(!any_of(as.character(weightby))) %>% # then we can drop the score since it will be nonsensical after the aggregation anyways group_by(label,seqnames,anno,class) %>% dplyr::summarise(across( where(is.numeric), .fns=sum )) %>% # sum up genewise feature counts ungroup() # From here we need to cast the motifcounts for the promoterregions, so that in the end we have one row per gene (instead of 1-2) motifdf_aggr <- motifdf_aggr %>% tidyr::pivot_wider(id_cols=c(label,seqnames,anno), names_from=class, values_from = !c(label,seqnames,anno, class), values_fill = 0) } else { motifdf_aggr <- motifdf %>% dplyr::select(!c(unselect_col,"name","baseMean","log2FoldChange","lfcSE","stat","pvalue","padj","gene_biotype","mgi_symbol")) %>% { if(weightby!=FALSE) mutate(., across(where(is.numeric), ~ (.x * get(weightby)))) else .} %>% # weight features by score dplyr::select(!any_of(as.character(weightby))) %>% # then we can drop the score since it will be nonsensical after the aggregation anyways group_by(label,seqnames,anno) %>% dplyr::summarise(across( where(is.numeric), .fns=sum )) %>% # sum up genewise feature counts ungroup() } } else { # loop for prox based assignments motifdf_aggr <- motifdf %>% dplyr::select(!c("name","baseMean","log2FoldChange","lfcSE","stat","pvalue","padj","gene_biotype")) %>% group_by(label,seqnames,anno) %>% dplyr::summarise(across( where(is.numeric), .fns=sum )) %>% # sum up genewise feature counts ungroup() } return(motifdf_aggr) } #------------function needs------------: # featurematrix and indexes to perform split of test vs trainval set # (trainval set is used for CV to find optimal regularization) #------------function will------------: # run cv.glmnet (as elastic net regression) and pick lambda.1se as regularization # determine model performance on test set # plots for raw counts of certain factors are currently turned of complete_GLM_analysis <- function(motifdf, trainvalidx, genenames){ # Create training subset for model development & testing set for model performance testing # make it based on chromosomes, instead of completely random # as in bpnet, we set aside chr 1,8 and 9 for testing #inTrain <- sort(sample(nrow(motifdata_abc_aggr), nrow(motifdata_abc_aggr)*0.75)) features_train <- motifdf[ trainvalidx, -c(1,2,3)] %>% as.matrix() features_test <- motifdf[ -trainvalidx, -c(1,2,3)] %>% as.matrix() targets_train <- motifdf[ trainvalidx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() targets_test <- motifdf[ -trainvalidx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() #------------------------ ## Elastic net #------------------------ cvfit_net <- glmnet::cv.glmnet(x=features_train, y=targets_train, family="binomial", type.measure = "auc", nfolds = 6, alpha=0.5) # Performance on training data targets_train_net.prob <- predict(cvfit_net, type="response", newx = features_train, s = 'lambda.min') pred_train_net <- ROCR::prediction(targets_train_net.prob[,1], targets_train) #only need the probabilities for 1's auc_ROCR_train_net <- ROCR::performance(pred_train_net, measure = "auc") #to assess AUC for model auc_ROCR_train_net@y.values[[1]] # Predict on unseen data targets_net.prob <- predict(cvfit_net, type="response", newx = features_test, s = 'lambda.min') pred_net <- ROCR::prediction(targets_net.prob[,1], targets_test) #only need the probabilities for 1's auc_ROCR_net <- ROCR::performance(pred_net, measure = "auc") #to assess AUC for model auc_ROCR_net@y.values[[1]] #------------------------ ## Model coefficients #------------------------ coef_df_net <- coerce_coef2df( coefficients(cvfit_net, s = 'lambda.min')) gg_net <- ggplot( data=coef_df_net %>% filter(abs(estimates)>0) )+ geom_bar(aes(x=reorder(names, -estimates), y=estimates), stat="identity")+ labs(title=" ",x="", y="Net coefs")+ #coord_flip()+ theme(axis.text.x = element_text(angle = 45, hjust=1), plot.margin = unit(c(0, 0, -6, 10), "points")) #------------------------ # ROC curves #------------------------ # calculate probabilities for TPR/FPR for predictions perf_train_net <- ROCR::performance(pred_train_net,"tpr","fpr") perf_net <- ROCR::performance(pred_net,"tpr","fpr") # plot ROC curve gg_ROC <- ggplot()+ geom_line( aes(x=perf_train_net@x.values[[1]], y=perf_train_net@y.values[[1]], colour = "train_net") ) + geom_line( aes(x=perf_net@x.values[[1]],y=perf_net@y.values[[1]], colour = "test_net") ) + geom_abline(intercept=0,slope=1, linetype=4, colour="grey")+ scale_colour_manual(values=c("darkblue", "blue"), name=" ", breaks=c("train_net", "test_net"), labels=c(paste("net train. AUC:",round( auc_ROCR_train_net@y.values[[1]], 2)) , paste("net test. AUC:", round( auc_ROCR_net@y.values[[1]], 2)) ) )+ labs(x="False positive rate", y="True positive rate")+ theme( legend.position=c(0.65,0.15) ) # add metric to global variable new_metrics = data.frame( net_train = auc_ROCR_train_net@y.values[[1]], net_test = auc_ROCR_net@y.values[[1]] ) full_panel <- ggpubr::ggarrange(gg_ROC,gg_net, nrow=2, labels=c("A","B"), heights = c(1,1)) results <- list() results[[1]] <- full_panel results[[2]] <- new_metrics results[[3]] <- coef_df_net results[[4]] <- cvfit_net return(results) } #------------------------------------------------ ## initialize object to track performance metrics #------------------------------------------------ AUC_metrics = data.frame( # parameter combination condition=character(), motifdata=character(), excludepromoters=character(), onlymax=character(), weight=character(), sepPromEnh=character(), # performance net_train = numeric(), net_test = numeric() ) net_model_coefs = list() #------------------------------------------------ ## run proximity based #------------------------------------------------ # This is independent of the ABC results (no need to loop through different assignment variations) print("Running GLM on prox-based assignments") my_rdsfile <- paste0(opt$outdir,"prox.rds") if(!file.exists(my_rdsfile)){ motifdata_aggr <- merge_motifdata_with_assignments(motifcounts_summitregion, assignment_summit_prox, contrast_DexLPSvLPS, maxonly="prox", excludepromoters=FALSE, weightby = FALSE, sepPromEnh = FALSE) motifdata_aggr_scaled <- motifdata_aggr %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) motifdata_aggr_tranval_idx <- motifdata_aggr_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) performance <- complete_GLM_analysis (motifdata_aggr_scaled, motifdata_aggr_tranval_idx, genenames=genenames) #------------------------ ## Raw counts #------------------------ nr3c1counts <- motifdata_aggr %>% group_by(label) %>% dplyr::count(NR3C1) %>% mutate(counts_fac = case_when(NR3C1>=1 ~ ">=1", NR3C1==0 ~ "0"), # aggregate it towards to top end counts_fac = factor(as.character(counts_fac), levels=c("0",">=1")) ) %>% group_by(counts_fac) %>% mutate(freq = n / sum(n)) %>% group_by(label,counts_fac) %>% dplyr::summarize(sum_n=sum(n), sum_freq=sum(freq)) gg_nr3c1counts <- ggplot(data=nr3c1counts , aes(x=counts_fac,y=sum_freq,fill=label)) + geom_bar(stat="identity", position="dodge", alpha=0.5)+ geom_text(aes(label=format(sum_freq, digits=2)), position = position_dodge(.9), vjust = -0.5, size = 3) + scale_fill_manual(values=c("0"="blue", "1"="red"), labels=c("DOWN","UP"))+ labs(x="NR3C1 matches", y="% genes", fill="")+ theme( plot.margin = unit(c(10, 6, 0, 10), "points") ) gg_nr3c1counts relcounts <- motifdata_aggr %>% group_by(label) %>% dplyr::count(REL) %>% mutate(counts_fac = case_when(REL>=1 ~ ">=1", REL==0 ~ "0"), # aggregate it towards to top end counts_fac = factor(as.character(counts_fac), levels=c("0",">=1")) ) %>% group_by(counts_fac) %>% mutate(freq = n / sum(n)) %>% group_by(label,counts_fac) %>% dplyr::summarize(sum_n=sum(n), sum_freq=sum(freq)) gg_relcounts <- ggplot(data=relcounts , aes(x=counts_fac,y=sum_freq,fill=label)) + geom_bar(stat="identity", position="dodge", alpha=0.5)+ geom_text(aes(label=format(sum_freq, digits=2)), position = position_dodge(.9), vjust = -0.5, size = 3) + scale_fill_manual(values=c("0"="blue", "1"="red"), labels=c("DOWN","UP"))+ labs(x="REL matches", y="% genes", fill="")+ theme( plot.margin = unit(c(10, 6, 0, 10), "points") ) gg_relcounts gg_rawcounts <- ggpubr::ggarrange( gg_nr3c1counts, gg_relcounts, ncol=2, common.legend = TRUE, legend="bottom") saveRDS(gg_rawcounts, paste0(opt$outdir,"gg_rawcounts.rds")) #------------------------ saveRDS(performance, file=my_rdsfile) } else { performance <- readRDS(my_rdsfile) } plot(performance[[1]]) AUC_metrics <- rbind(AUC_metrics, "prox" = c( condition="dexlps", motifdata="motifcounts_summitregion", excludepromoters=FALSE, onlymax="prox", weight=FALSE, sepPromEnh=FALSE, performance[[2]] ) ) net_model_coefs[["prox"]] <- performance[[3]] ## Raw counts gg_firstgene <- ggplot() + geom_bar(data=motifdata_aggr , aes(NR3C1,fill=label), position="dodge", alpha=0.5)+ scale_fill_manual(values=c("0"="blue", "1"="red"), labels=c("DOWN","UP"))+ labs(x="NR3C1", y="counts", fill="")+ theme( plot.margin = unit(c(10, 6, 0, 10), "points") ) gg_firstgene #------------------------------------------------ ## run example #------------------------------------------------ # motifdata_aggr <- merge_motifdata_with_assignments( # motifcounts_summitregion, # assignment_summitregion_abc_dexlps, # contrast_DexLPSvLPS, # maxonly=FALSE, # excludepromoters=FALSE, # weightby=FALSE, # sepPromEnh=FALSE) # # motifdata_aggr_tranval_idx <- motifdata_aggr %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) # performance <- complete_GLM_analysis (motifdata_aggr, motifdata_aggr_tranval_idx, genenames=genenames) # plot(performance[[1]]) # AUC_metrics <- rbind(AUC_metrics, # "test" = c(condition="dexlps", # motifdata="motifcounts_peak", # onlymax=FALSE, # excludepromoters=FALSE, # weight=FALSE, # sepPromEnh=FALSE, # performance[[2]])) # net_model_coefs[["test"]] <- performance[[3]] # # gg_firstgene <- ggplot() + # geom_bar(data=performance[[5]] , aes(NR3C1,fill=label), position="dodge", alpha=0.5)+ # scale_fill_manual(values=c("0"="blue", "1"="red"), # labels=c("DOWN","UP"))+ # labs(x="NR3C1", y="counts", fill="")+ # theme( # plot.margin = unit(c(10, 6, 0, 10), "points") # ) # gg_firstgene #------------------------------------------------ ## peakregion_counts - proxbased assignment ## peakregion_counts - abcbased assignment (both conditions) ## enhancerregion_counts - abcbased assignment (2 conditions) #------------------------------------------------ # rewrite to assign values for both conditions and then use those to compute the difference as well not_all_na <- function(x) any(!is.na(x)) for (motifdata in c("motifcounts_abcregion","motifcounts_summitregion")){ for(excludepromoters in c(FALSE,"all","onlyNONself")){ for (onlymax in c(TRUE,FALSE)){ for (sepPromEnh in c(TRUE,FALSE)){ for (weight in c(FALSE, "abcscore")){ # set assignments fitting for the input data if(motifdata=="motifcounts_abcregion"){ assignment_dexlps <- assignment_abcregion_dexlps assignment_lps <- assignment_abcregion_lps motifcounts_dexlps <- motifcounts_abcregion_dexlps motifcounts_lps <- motifcounts_abcregion_lps } else if (motifdata=="motifcounts_summitregion" ){ # in this case the motifcounts for the 2 conditions are the same, but there assignments differ assignment_dexlps <- assignment_summits_abcregion_dexlps assignment_lps <- assignment_summits_abcregion_lps motifcounts_dexlps <- motifcounts_summitregion motifcounts_lps <- motifcounts_summitregion } else {break} #-----------------------DEXLPS------------------------------ featurematrix_dexlps <- merge_motifdata_with_assignments(motifcounts_dexlps, assignment_dexlps, contrast_DexLPSvLPS, weightby=weight, maxonly=onlymax, excludepromoters=excludepromoters, sepPromEnh=sepPromEnh) # remove columns that are NA after scaling featurematrix_dexlps_scaled <- featurematrix_dexlps %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) %>% dplyr::select(where(not_all_na)) #----- motifdata_aggr_tranval_idx <- featurematrix_dexlps_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) modelname <- paste(motifdata,"condition_dexlps_exclprom",excludepromoters,"onlymax",onlymax,"sepPromEnh",sepPromEnh,"weight",weight, sep="_") print(modelname) my_rdsfile <- here( paste0(opt$outdir, modelname,".rds")) if(!file.exists(my_rdsfile)){ performance <- complete_GLM_analysis (featurematrix_dexlps_scaled, motifdata_aggr_tranval_idx, genenames=genenames) saveRDS(performance,file=my_rdsfile) } else { performance <- readRDS(my_rdsfile) } AUC_metrics <- rbind(AUC_metrics, c( condition="dexlps", motifdata=motifdata, excludepromoters=excludepromoters, onlymax=onlymax, weight=weight, sepPromEnh=sepPromEnh, performance[[2]] ) )%>% magrittr::set_rownames(c(rownames(AUC_metrics),modelname)) net_model_coefs[[modelname]] <- performance[[3]] #--------------------------LPS--------------------------- featurematrix_lps <- merge_motifdata_with_assignments(motifcounts_lps, assignment_lps, contrast_DexLPSvLPS, weightby=weight, maxonly=onlymax, excludepromoters=excludepromoters, sepPromEnh=sepPromEnh) featurematrix_lps_scaled <- featurematrix_lps %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) %>% dplyr::select(where(not_all_na)) #----- motifdata_aggr_tranval_idx <- featurematrix_lps_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) modelname <- paste(motifdata,"condition_lps_exclprom",excludepromoters,"onlymax",onlymax,"sepPromEnh",sepPromEnh,"weight",weight, sep="_") print(modelname) my_rdsfile <- here(paste0(opt$outdir, modelname,".rds")) if(!file.exists(my_rdsfile)){ performance <- complete_GLM_analysis (featurematrix_lps_scaled, motifdata_aggr_tranval_idx, genenames=genenames) saveRDS(performance,file=my_rdsfile) } else { performance <- readRDS(my_rdsfile) } AUC_metrics <- rbind(AUC_metrics, c( condition="lps", motifdata=motifdata, excludepromoters=excludepromoters, onlymax=onlymax, weight=weight, sepPromEnh=sepPromEnh, performance[[2]] ) )%>% magrittr::set_rownames(c(rownames(AUC_metrics),modelname)) net_model_coefs[[modelname]] <- performance[[3]] #-----------------------DIFFERENCE------------------------------ # they contain the same motifs, but not the same genes. # doublecheck that all columnnames are identical table(colnames(featurematrix_dexlps) == colnames(featurematrix_lps)) # motifcounts missing in one condition should be set to 0 merged_featurematrix <- merge(featurematrix_dexlps, featurematrix_lps, by=c("anno","label","seqnames"), all=TRUE) # replace missing counts in one of the conditions with 0 merged_featurematrix[is.na(merged_featurematrix)] <- 0 # use dataframe suffix to grab respective columns featurematrix_diff <- merged_featurematrix[grep(".x$",colnames(merged_featurematrix))] - merged_featurematrix[grep(".y$",colnames(merged_featurematrix))] # tidy up column names colnames(featurematrix_diff) <- gsub(".x$","", colnames(featurematrix_diff)) # add first 3 columns back after computing the difference of the counts featurematrix_diff <- cbind(merged_featurematrix[1:3],featurematrix_diff) featurematrix_diff_scaled <- featurematrix_diff %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) %>% dplyr::select(where(not_all_na)) # run GLM and look at performance featurematrix_diff_tranval_idx <- featurematrix_diff_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) modelname <- paste(motifdata,"condition_DexLPS-LPS_exclprom",excludepromoters,"onlymax",onlymax,"sepPromEnh",sepPromEnh,"weight",weight, sep="_") print(modelname) my_rdsfile <- here(paste0(opt$outdir, modelname,".rds")) if(!file.exists(my_rdsfile)){ performance <- complete_GLM_analysis (featurematrix_diff_scaled, featurematrix_diff_tranval_idx, genenames=genenames) saveRDS(performance,file=my_rdsfile) } else { performance <- readRDS(my_rdsfile) } AUC_metrics <- rbind(AUC_metrics, c( condition="DexLPS-LPS", motifdata=motifdata, excludepromoters=excludepromoters, onlymax=onlymax, weight=weight, sepPromEnh=sepPromEnh, performance[[2]] ) )%>% magrittr::set_rownames(c(rownames(AUC_metrics),modelname)) net_model_coefs[[modelname]] <- performance[[3]] } } } } } #------------------------------------------------ ## put model coefs into dataframes #------------------------------------------------ motifnames <- colnames(motifcounts_summitregion)[3:ncol(motifcounts_summitregion)] # make empty dataframe with all motifs model_coefs_joint <- data.frame(featurename=motifnames) # when we treat promoters and enhancers separately, the featurenames differ model_coefs_sep <- data.frame(featurename= c(paste(motifnames,"genic",sep="_"), paste(motifnames,"promoter",sep="_"), paste(motifnames,"intergenic",sep="_")) ) for (model in names(net_model_coefs)){ if (grepl("sepPromEnh_FALSE",model)){ model_coefs_joint <- merge(model_coefs_joint, net_model_coefs[[model]], by.x="featurename", by.y="names", all=TRUE) %>% dplyr::rename(!! model := "estimates") } else if (model=="prox"){ model_coefs_joint <- merge(model_coefs_joint, net_model_coefs[[model]], by.x="featurename", by.y="names", all=TRUE) %>% dplyr::rename(!! model := "estimates") } else if (grepl("sepPromEnh_TRUE",model)){ model_coefs_sep <- merge(model_coefs_sep, net_model_coefs[[model]], by.x="featurename", by.y="names", all=TRUE) %>% dplyr::rename(!! model := "estimates") } else { stop("The modelnames don't match the expected pattern.") } } #------------------------------------------------ ## save data for plotting #------------------------------------------------ saveRDS (model_coefs_joint, paste0(opt$outdir, "model_coefs_joint.rds")) saveRDS (model_coefs_sep, paste0(opt$outdir, "model_coefs_sep.rds")) saveRDS (AUC_metrics, paste0(opt$outdir, "AUC_metrics.rds")) #------------------------------------------------ sessionInfo() |
R
ggplot2
dplyr
org.Mm.eg.db
optparse
here
ComplexHeatmap
FiMO
plyranges
From
line
1
of
scripts/abc_predictions.r
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--ABC_DexLPS_all"), type="character", help="Path to file with all ABC predictions passing 0.02 (in DexLPS condition)"), make_option(c("--ABC_LPS_all"), type="character", help="Path to file with all ABC predictions passing 0.02 (in LPS condition)"), make_option(c("--contrast_DexVSDexLPS"), type="character", help="Path to annotated tsv file of DeSeq2 contrast of DexLPS vs LPS"), make_option(c("--chipseq_summits"), type="character", help="Path to summit file of IDR peaks"), make_option(c("--igv"), type="character", help="Path to IGV snapshot of genomic locus (created manually)") ) opt <- parse_args(OptionParser(option_list=option_list)) options(stringsAsFactors = FALSE) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(org.Mm.eg.db, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(DESeq2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ComplexHeatmap, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(TxDb.Mmusculus.UCSC.mm10.knownGene, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(topGO, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggExtra, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=8, family = "ArialMT", colour="black"), title=element_text(size=10, family="ArialMT", colour="black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text = element_text(size=8, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black")) #---------------------------------------------------------------------- ##--NOTE: example code for additional plots can be found in abc_visualizations.Rmd #---------------------------------------------------------------------- #---------------------------------------------------------------------- ##----------- load data #---------------------------------------------------------------------- chipseq_summits <- rtracklayer::import.bed(opt$chipseq_summits) contrast_DexLPSvLPS <- read.delim(opt$contrast_DexVSDexLPS) # rename first column for merge later colnames(contrast_DexLPSvLPS)[1] <- "EnsemblID" ABC_DexLPS_all <- read.delim(opt$ABC_DexLPS_all) ABC_LPS_all <- read.delim(opt$ABC_LPS_all) # remove version number from ensembl IDs ABC_DexLPS_all$TargetGene = gsub("\\.[0-9]*$","",ABC_DexLPS_all$TargetGene) ABC_LPS_all$TargetGene = gsub("\\.[0-9]*$","",ABC_LPS_all$TargetGene) #---------------------------------------------------------------------- ##----------- data exploration #---------------------------------------------------------------------- # What do the distributions of ABC scores look like ggplot()+ geom_density(data=ABC_DexLPS_all %>% filter(!class=="promoter"), aes(x=ABC.Score, colour="DexLPS"))+ geom_density(data=ABC_LPS_all %>% filter(!class=="promoter"), aes(x=ABC.Score, colour="LPS"))+ scale_colour_manual("", breaks = c("DexLPS", "LPS"), values = c("#339966", "#0066CC")) # How do the ABC scores of promoters compare to those of enhancers? ggplot()+ geom_density(data=ABC_DexLPS_all %>% filter(class=="promoter"), aes(x=ABC.Score, colour="promoter"))+ geom_density(data=ABC_DexLPS_all %>% filter(class=="genic"), aes(x=ABC.Score, colour="genic"))+ geom_density(data=ABC_DexLPS_all %>% filter(class=="intergenic"), aes(x=ABC.Score, colour="intergenic"))+ scale_colour_manual("", breaks = c("promoter", "genic", "intergenic"), values = c("red", "green", "blue")) # What distance is used to define sth as promoter? ggplot()+ geom_density(data=ABC_DexLPS_all, aes(x=distance, colour=class))+ scale_x_continuous(trans = "log10")+ scale_colour_manual("", breaks = c("promoter", "genic", "intergenic"), values = c("red", "green", "blue")) ggplot()+ geom_density(data=ABC_DexLPS_all %>% filter(class=="promoter"), aes(x=distance, colour=isSelfPromoter))+ scale_x_continuous(trans = "log10")+ scale_colour_manual("isSelfPromoter", breaks = c("True", "False"), values = c("red", "blue"))+ labs(title="All promoter regions, split by whether they're promoters for their target gene")+ theme( plot.title = element_text(size=12) ) #---------------------------------------------------------------------- # ------ ------ Number of enhancers per gene #---------------------------------------------------------------------- # Are most genes regulated by a single enhancer region or by multiple? How many does the ABC model assign per gene? plot_enhancers_per_gene <- function(){ DexLPS_TargetGeneCounts <- plyr::count(ABC_DexLPS_all %>% filter(!class=="promoter"), vars="TargetGene") LPS_TargetGeneCounts <- plyr::count(ABC_LPS_all %>% filter(!class=="promoter"), vars="TargetGene") gg <- ggplot()+ geom_histogram(data=DexLPS_TargetGeneCounts, aes(x=freq, fill="DexLPS", colour="DexLPS"), binwidth = 1, alpha=0.2)+ geom_histogram(data=LPS_TargetGeneCounts, aes(x=freq, fill="LPS", colour="LPS"), binwidth = 1, alpha=0.2)+ scale_fill_manual("", breaks = c("DexLPS", "LPS"), values = c("#339966", "#0066CC")) + scale_colour_manual("", breaks = c("DexLPS", "LPS"), values = c("#339966", "#0066CC")) + guides(color = "none")+ theme(legend.position = c(0.8,0.8))+ labs(title=" ", x="# of enhancers per gene", y="counts") return(gg) } gg_enh_per_gene <- plot_enhancers_per_gene() gg_enh_per_gene print("We find an average of") plyr::count(ABC_DexLPS_all %>% filter(!class=="promoter"), vars="TargetGene") %>% pull(freq) %>% mean() print("enhancers per gene for the DexLPS condition and") plyr::count(ABC_LPS_all %>% filter(!class=="promoter"), vars="TargetGene") %>% pull(freq) %>% mean() print("in LPS.") #---------------------------------------------------------------------- # ------ ------ Number of genes per enhancer #---------------------------------------------------------------------- # ABC allows for multiple assignments (an individual enhancer assigned to more than one promoter) plot_genes_per_enhancer <- function(){ DexLPS_EnhancerCounts <- plyr::count(ABC_DexLPS_all %>% filter(!class=="promoter"),vars="name") LPS_EnhancerCounts <- plyr::count(ABC_LPS_all %>% filter(!class=="promoter"),vars="name") gg_full <- ggplot()+ geom_histogram(data=DexLPS_EnhancerCounts, aes(x=freq, fill="DexLPS",colour="DexLPS"), binwidth = 1, alpha=0.2)+ geom_histogram(data=LPS_EnhancerCounts, aes(x=freq, fill="LPS",colour="LPS"), binwidth = 1, alpha=0.2)+ scale_fill_manual("", breaks = c("DexLPS", "LPS"), values = c("#339966", "#0066CC")) + scale_colour_manual("", breaks = c("DexLPS", "LPS"), values = c("#339966", "#0066CC")) + guides(color = FALSE)+ labs(title="Number of genes an individual enhancer is assigned to", x="# of genes per enhancer", y="counts") plot(gg_full) gg_zoom <- ggplot()+ geom_histogram(data=DexLPS_EnhancerCounts, aes(x=freq, fill="DexLPS",colour="DexLPS"), binwidth = 1, alpha=0.2)+ geom_histogram(data=LPS_EnhancerCounts, aes(x=freq, fill="LPS",colour="LPS"), binwidth = 1, alpha=0.2)+ scale_fill_manual("", breaks = c("DexLPS", "LPS"), values = c("#339966", "#0066CC")) + scale_colour_manual("", breaks = c("DexLPS", "LPS"), values = c("#339966", "#0066CC")) + xlim(c(0,8))+ guides(color = FALSE)+ theme(legend.position = c(0.8,0.8))+ labs(title=" ", x="# of genes per enhancer", y="counts") plot(gg_zoom) print("We find an average of") DexLPS_EnhancerCounts %>% pull(freq) %>% mean() %>% print() print("genes per enhancer in the DexLPS condition and") LPS_EnhancerCounts %>% pull(freq) %>% mean() %>% print() print("in LPS.") return(gg_zoom) } gg_genes_per_enhancer_zoom <- plot_genes_per_enhancer() #---------------------------------------------------------------------- # ------ Comparing ABC Scores (including promoter regions) between the conditions #---------------------------------------------------------------------- # merge regions from both conditions but also keep those that are only in one # # add info no whether the target genes are DE or now and what direction they're changed LPS_GR<- plyranges::as_granges(ABC_LPS_all, seqnames=chr) DexLPS_GR <- plyranges::as_granges(ABC_DexLPS_all, seqnames=chr) #---------------------------------------------------------------------- merge_LPSandDexLPS_ABCdata <- function(ABC_LPS_all_GR, ABC_DexLPS_all_GR){ ABC_LPS_all_unique <- unique(ABC_LPS_all_GR) ABC_DexLPS_all_unique <- unique(ABC_DexLPS_all_GR) # and then figure out which ones overlap with one another overlap_ABC_all_unique <- ChIPpeakAnno::findOverlapsOfPeaks (ABC_LPS_all_unique, ABC_DexLPS_all_unique, connectedPeaks = "keepAll", ignore.strand = TRUE) # we coerce the info of overlapping regions into a dataframe overlap_ABC_all_unique_df <- as.data.frame(overlap_ABC_all_unique$overlappingPeaks) # and clean up column names colnames(overlap_ABC_all_unique_df)<- gsub("ABC_LPS_all_unique...ABC_DexLPS_all_unique.","", colnames(overlap_ABC_all_unique_df)) colnames(overlap_ABC_all_unique_df)<- gsub(".1", "", colnames(overlap_ABC_all_unique_df)) colnames(overlap_ABC_all_unique_df)<- paste( c(rep("LPS_",27),rep("DexLPS_",27),rep("",2)), colnames(overlap_ABC_all_unique_df), sep="") # we use this overlap to define a new variable that holds the info of which regions of the conditions form a pair overlap_ABC_all_unique_df$pairID <- paste("pair", seq(1, nrow(overlap_ABC_all_unique_df)), sep="" ) #-----------------for DexLPS------------------- # merge the pairID onto the original dataframes ABC_DexLPS_all_wpairID <- merge(ABC_DexLPS_all, overlap_ABC_all_unique_df[,c("DexLPS_name","pairID")], by.x="name", by.y="DexLPS_name", all.x=TRUE) # for those that don't correspond to a pair, we assign condition and name as pairID ABC_DexLPS_all_wpairID <- ABC_DexLPS_all_wpairID %>% mutate(pairID = case_when(is.na(pairID) ~ paste("DexLPS",name, sep="_"), TRUE ~ pairID) ) # combination of pairID and gene they're assigned to forms the assignment variable. # we will use this later to plot values from the 2 conditions, that have the same assignment ABC_DexLPS_all_wpairID$assignment <- paste(ABC_DexLPS_all_wpairID$pairID, ABC_DexLPS_all_wpairID$TargetGene, sep="_") #-----------------for LPS------------------- # merge the pairID onto the original dataframes ABC_LPS_all_wpairID <- merge(ABC_LPS_all, overlap_ABC_all_unique_df[,c("LPS_name","pairID")], by.x="name", by.y="LPS_name", all.x=TRUE) # for those that don't correspond to a pair, we assign condition and name as pairID ABC_LPS_all_wpairID <- ABC_LPS_all_wpairID %>% mutate(pairID = case_when(is.na(pairID) ~ paste("LPS",name, sep="_"), TRUE ~ pairID) ) ABC_LPS_all_wpairID$assignment <- paste(ABC_LPS_all_wpairID$pairID, ABC_LPS_all_wpairID$TargetGene, sep="_") merged_conditions <- merge(ABC_DexLPS_all_wpairID, ABC_LPS_all_wpairID, by="assignment", all=TRUE, suffixes = c(".DexLPS", ".LPS")) merged_conditions <- merged_conditions %>% mutate(TargetGene = case_when(TargetGene.DexLPS==TargetGene.LPS ~ TargetGene.DexLPS, is.na(TargetGene.DexLPS) ~ TargetGene.LPS, is.na(TargetGene.LPS) ~ TargetGene.DexLPS, TRUE ~ NA_character_) ) return(merged_conditions) } merged_conditions_ABC <- merge_LPSandDexLPS_ABCdata(LPS_GR,DexLPS_GR) #---------------------------------------------------------------------- # ------ merge on gene expression results #---------------------------------------------------------------------- # merge the RNAseq results to use as colour later merged_conditions_ABC <- merge(merged_conditions_ABC, contrast_DexLPSvLPS, by.x="TargetGene", by.y="EnsemblID") merged_conditions_ABC <- merged_conditions_ABC %>% mutate(change=case_when(padj<0.05 & log2FoldChange>0.58 ~ "up", padj<0.05 & log2FoldChange<(-0.58) ~ "down", TRUE ~ "no change")) # replace the na with 0 merged_conditions_ABCnomissing <- merged_conditions_ABC %>% mutate(ABC.Score.DexLPS = case_when(is.na(ABC.Score.DexLPS) ~ 0, TRUE ~ ABC.Score.DexLPS) ) %>% mutate(ABC.Score.LPS = case_when(is.na(ABC.Score.LPS) ~ 0, TRUE ~ ABC.Score.LPS) ) cor_ABCscores <- cor(merged_conditions_ABCnomissing$ABC.Score.DexLPS, merged_conditions_ABCnomissing$ABC.Score.LPS) gg_merged_conditions_ABCnomissing <- ggplot(merged_conditions_ABCnomissing)+ geom_point(aes(x=ABC.Score.DexLPS, y=ABC.Score.LPS, fill=log2FoldChange), alpha=0.2, size=1, stroke = 0, shape=21)+ scale_fill_gradient(low="blue", high="red")+ ylim(c(0,0.8))+ xlim(c(0,0.8))+ geom_text(x=0.2, y=0.75, label=paste("r = ",format(cor_ABCscores, digits = 2)))+ labs(fill="log2FC", x="ABC score Dex+LPS", y="ABC score LPS")+ theme(legend.position = "top") gg_merged_conditions_ABCnomissing #---------------------------------------------------------------------- # ------ plot foldchange in ABC score vs foldchange in expression #---------------------------------------------------------------------- cor_ABCdiff_log2FC <- with(merged_conditions_ABCnomissing %>% filter(change!="no change"), cor((ABC.Score.DexLPS - ABC.Score.LPS),log2FoldChange)) gg_ABCdiff_log2FC <- ggplot(merged_conditions_ABCnomissing %>% filter(change!="no change"), aes(x=(ABC.Score.DexLPS - ABC.Score.LPS), y=log2FoldChange)) + geom_hex(bins=70)+ viridis::scale_fill_viridis()+ geom_smooth(method = "lm", colour="black")+ geom_text(x=0.6, y=5, label=paste("r = ",format(cor_ABCdiff_log2FC, digits = 2)))+ labs(x="ABC score Dex+LPS - ABC score LPS", y="expression log2FC(Dex+LPS / LPS)")+ theme(legend.position = "top") #---------------------------------------------------------------------- # ------ ABC score by peak presence #---------------------------------------------------------------------- plot_ABC_with_marginals <- function(whatchange){ gg <- ggplot(data= merged_conditions_ABCnomissing %>% filter(change==whatchange) , aes(x=ABC.Score.DexLPS, y=ABC.Score.LPS ) )+ geom_point( aes( fill=whatchange) , size=1, alpha=1,stroke = 0, shape=21) + geom_abline(slope = 1)+ ylim(c(0,0.8))+ xlim(c(0,0.8))+ labs(x="ABC score Dex+LPS", y="ABC score LPS")+ scale_fill_manual( breaks=c("up","no change","down"), values=c("red","black","blue") )+ theme(legend.position = "none") gg_marg <- ggExtra::ggMarginal(gg, type="histogram", size=2, xparams = list(bins=85), yparams = list(bins=85)) return(gg_marg) } ggplot_ABC_with_marginals_up <- plot_ABC_with_marginals("up") ggplot_ABC_with_marginals_up ggplot_ABC_with_marginals_down <-plot_ABC_with_marginals("down") ggplot_ABC_with_marginals_down #---------------------------------------------------------------------- # ------ show marginals as bar plot #---------------------------------------------------------------------- gg_marginals_dexlps <- ggplot()+ geom_histogram(data=merged_conditions_ABCnomissing %>% filter(change=="up"), aes(x=ABC.Score.DexLPS, fill="up"), alpha=0.2, bins=85)+ geom_histogram(data=merged_conditions_ABCnomissing %>% filter(change=="down"), aes(x=ABC.Score.DexLPS, fill="down"), alpha=0.2, bins=85)+ scale_fill_manual( breaks=c("up","down"), values=c("red","blue") )+ xlim(c(NA,0.3))+ labs(fill="gene expression change",x="ABC score DexLPS", y="ABC score LPS")+ theme(legend.position = c(0.7,0.7)) gg_marginals_lps <- ggplot()+ geom_histogram(data=merged_conditions_ABCnomissing %>% filter(change=="up"), aes(x=ABC.Score.LPS, fill="up"), alpha=0.2, bins=85)+ geom_histogram(data=merged_conditions_ABCnomissing %>% filter(change=="down"), aes(x=ABC.Score.LPS, fill="down"), alpha=0.2, bins=85)+ scale_fill_manual( breaks=c("up","down"), values=c("red","blue") )+ xlim(c(NA,0.3))+ labs(fill="gene expression change",x="ABC score DexLPS", y="ABC score LPS")+ theme(legend.position = c(0.7,0.7)) #---------------------------------------------------------------------- # ------ ABC score by peak presence #---------------------------------------------------------------------- # Is there a difference in ABC score between the enhancers overlapping with a GR summit vs the ones that don't? DexLPS_GR_summits_overlap <- IRanges::findOverlaps(DexLPS_GR,chipseq_summits) #(query,subject) DexLPS_GR$haspeakID <- FALSE DexLPS_GR$haspeakID[queryHits(DexLPS_GR_summits_overlap)] <- TRUE plot_ABCscore_ofDEgenes_byGRoverlap <- function(){ DE_ENSEMBL <- contrast_DexLPSvLPS %>% filter(padj<0.05 & abs(log2FoldChange)>0.58 ) %>% pull(EnsemblID) DexLPS_enhancers_DEsubset <- as.data.frame(DexLPS_GR) %>% filter(TargetGene %in% DE_ENSEMBL) %>% filter(!class=="promoter") DexLPS_enhancers_DEsubset %>% filter(haspeakID==TRUE) %>% dplyr::pull(ABC.Score) %>% mean() %>% print() DexLPS_enhancers_DEsubset %>% filter(haspeakID==FALSE) %>% dplyr::pull(ABC.Score) %>% mean() %>% print() # show distribution for the two groups gg1 <- ggplot()+ geom_density(data=DexLPS_enhancers_DEsubset %>% filter(haspeakID==TRUE), aes(x=ABC.Score, colour="haspeakID", fill="haspeakID"), alpha=0.3)+ geom_density(data=DexLPS_enhancers_DEsubset %>% filter(haspeakID==FALSE), aes(x=ABC.Score, colour="nopeakID", fill="nopeakID"), alpha=0.3)+ scale_x_continuous(trans = "log2")+ scale_colour_manual("", labels = c("has GR peak", "no GR peak"), breaks = c("haspeakID", "nopeakID"), values = c("darkmagenta", "cadetblue"))+ scale_fill_manual("", labels = c("has GR peak", "no GR peak"), breaks = c("haspeakID", "nopeakID"), values = c("darkmagenta", "cadetblue"))+ guides(colour="none")+ #xlim(c(-3,0))+ labs(title=" ", x="ABC score Dex+LPS", y="density")+ theme( axis.text = element_text(size=12, family="ArialMT", colour="black"), axis.title = element_text(size=12, family="ArialMT", colour="black"), axis.text.x = element_text(angle = 45, vjust = 1, hjust=1), legend.position = c(0.8,0.8) ) plot(gg1) gg2 <- DexLPS_enhancers_DEsubset %>% ggplot( aes(x=hic_contact_pl_scaled_adj, y=log2(activity_base))) + geom_hex()+ facet_wrap(~haspeakID, labeller="label_both")+ viridis::scale_fill_viridis()+ labs(title = " ", x="Hi-C contact", y="log2(base activity)") plot(gg2) all_plots=list() all_plots[[1]] <- gg1 all_plots[[2]] <- gg2 return(all_plots) } gg_ABCscore_ofDEgenes_byGRoverlap <- plot_ABCscore_ofDEgenes_byGRoverlap() #---------------------------------------------------------------------- # ------ load IGV plot #---------------------------------------------------------------------- igv <- png::readPNG( opt$igv ) gg_igv <- ggplot() + ggpubr::background_image(igv) + # so it doesn't get squished #coord_fixed()+ # This ensures that the image leaves some space at the edges theme(plot.margin = margin(t=0, l=0, r=0, b=0, unit = "cm"), axis.line = element_blank()) #---------------------------------------------------------------------- # ------ put figure panel together #---------------------------------------------------------------------- # save giant files separately ggsave(here("./results/current/Figures/Figure_abcresults_3B.bmp"), gg_merged_conditions_ABCnomissing, width=75, height=100, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_abcresults_3B.png"), gg_merged_conditions_ABCnomissing, width=75, height=100, units="mm", bg="white") gg_placeholder <- ggplot() + theme(plot.margin = margin(t=0, l=0, r=0, b=0, unit = "cm"), axis.line = element_blank()) # first row gg_r1c1 <- ggpubr::ggarrange(gg_enh_per_gene, gg_genes_per_enhancer_zoom, labels = c("A", NA), ncol = 1, nrow = 2, heights = c(1,1)) gg_r1c2 <- ggpubr::ggarrange(gg_placeholder, ggplot_ABC_with_marginals_up, ggplot_ABC_with_marginals_down, labels = c("B", "C", NA), ncol = 3, nrow=1, widths=c(1,1,1)) gg_r1 <- ggpubr::ggarrange(gg_r1c1, gg_r1c2, labels = c(NA, NA), ncol = 2, nrow=1, widths=c(1,3)) # second row gg_r2 <- ggpubr::ggarrange(gg_ABCdiff_log2FC, gg_ABCscore_ofDEgenes_byGRoverlap[[1]], gg_ABCscore_ofDEgenes_byGRoverlap[[2]], labels = c("D","E", "F"), ncol = 3, nrow=1, widths=c(1,1,1.5)) gg_tophalf <- ggpubr::ggarrange(gg_r1, gg_r2, nrow=2, heights =c(1,1)) gg_tophalf full_panel <- ggpubr::ggarrange(gg_tophalf, gg_igv, labels=c(NA,"G"), nrow=2, heights = c(1.8,1)) #full_panel ggsave(here("./results/current/Figures/Figure_abcresults.png"), full_panel, width=300, height=300, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_abcresults.pdf"), full_panel, width=300, height=300, units="mm", bg="white") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--bw_DexLPS_h3k27ac"), type="character", help="Path to H3K27ac bigwig file from DexLPS condition"), make_option(c("--bw_LPS_h3k27ac"), type="character", help="Path to H3K27ac bigwig file from LPS condition"), make_option(c("--counts_h3k27ac"), type="character", help="Path to file with adjusted libsize (number of reads in bam file overlappign joint peak universe)"), make_option(c( "--bw_DexLPS_atac"), type="character", help="Path to ATAC bigwig file from DexLPS condition"), make_option(c("--bw_LPS_atac"), type="character", help="Path to ATAC bigwig file from LPS condition"), make_option(c("--counts_atac"), type="character", help="Path to file with adjusted libsize (number of reads in bam file overlappign joint peak universe)"), make_option(c("--gtf"), type="character", help="Path to gtf file of genomic reference") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(rtracklayer, warn.conflicts=F, quietly=T)) #----------------------------------------- #------- H3K27ac #----------------------------------------- bw_DexLPS_h3k27ac <- import.bw( opt$bw_DexLPS_h3k27ac ) bw_LPS_h3k27ac <- import.bw( opt$bw_LPS_h3k27ac ) counts_h3k27ac <- read.table(opt$counts_h3k27ac) # normalization is important in this case as seen by the ratio as.numeric(counts_h3k27ac[2,1]) / as.numeric(counts_h3k27ac[4,1]) # scale by "adjusted" library size (number of reads overlapping peak universe) score(bw_DexLPS_h3k27ac) <- score(bw_DexLPS_h3k27ac) / ( as.numeric(counts_h3k27ac[2,1]) / 10^6) score(bw_LPS_h3k27ac) <- score(bw_LPS_h3k27ac) / ( as.numeric(counts_h3k27ac[4,1]) / 10^6) # reexport the normalized tracks export.bw(bw_DexLPS_h3k27ac, here("./results/current/ChIP/H3K27ac/bw/DexLPS_histone_H3K27ac_PE_merged_GRCm38_libnorm.bw")) export.bw(bw_LPS_h3k27ac, here("./results/current/ChIP/H3K27ac/bw/LPS_histone_H3K27ac_PE_merged_GRCm38_libnorm.bw")) rm(list=c("bw_DexLPS_h3k27ac", "bw_LPS_h3k27ac")) #----------------------------------------- #------- ATAC #----------------------------------------- bw_DexLPS_atac <- import.bw( opt$bw_DexLPS_atac ) bw_LPS_atac <- import.bw( opt$bw_LPS_atac ) counts_atac <- read.table(opt$counts_atac) # normalization is NOT so important in this case as seen by the ratio as.numeric(counts_atac[2,1]) / as.numeric(counts_atac[4,1]) # scale by "adjusted" library size (number of reads overlapping peak universe) score(bw_DexLPS_atac) <- score(bw_DexLPS_atac) / ( as.numeric(counts_atac[2,1]) / 10^6) score(bw_LPS_atac) <- score(bw_LPS_atac) / ( as.numeric(counts_atac[4,1]) / 10^6) # reexport the normalized tracks export.bw(bw_DexLPS_atac, here("./results/current/atacseq/bw/merged_DexLPS_GRCm38_libnorm.bw")) export.bw(bw_LPS_atac, here("./results/current/atacseq/bw/merged_LPS_GRCm38_libnorm.bw")) #----------------------------------------- #------- reference gtf file for IGV #----------------------------------------- #import gtf file with rtracklayer gtf <- rtracklayer::import(opt$gtf) gtf_nogene <- gtf %>% filter(type!="gene") rtracklayer::export(gtf_nogene, here("./data/current/Mus_musculus.GRCm38.100_ftrnogene.gtf")) rtracklayer::export.gff3(gtf_nogene, here("./data/current/Mus_musculus.GRCm38.100_ftrnogene.gff3")) |
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("-c", "--counts"), type="character", help="Path to count matrix"), make_option(c("-a", "--annotation"), type="character", help="Path to metadata with annotation"), make_option(c("-o", "--outdir"), type="character", help="Path to output directory"), make_option(c("-k", "--biomart_ensembl_version"), type="numeric", help="version of ensembl biomart to use for ensembl2mgi gene annotation")) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(edgeR, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(DESeq2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(biomaRt, warn.conflicts=F, quietly=T)) dir.create(opt$outdir) #------------------------------------------------- #---------Load data #------------------------------------------------- # Let's start by loading the count matrix holding the gene counts for all samples and by making a dataframe holding metadata such as treatment conditions. counts <- as.matrix(read.table(opt$counts, quote = "\"", header=TRUE, row.names = 1 )) anno <- read.delim(opt$annotation) # parse relevant info from treatment variable anno$treat <- anno$treatment %>% gsub(";4sU", "", .) %>% gsub("LPS, Dex", "LPS_Dex", .) %>% as.factor() %>% relevel(treat, ref = "Vehicle") # parse total vs nascent from Foreign.ID anno$group [ grepl('total|Total', anno$Foreign.ID) ] <- "total" anno$group [ grepl('nascent', anno$Foreign.ID) ] <- "nascent" # making sure anno is sorted to match column order in counts matrix anno <- anno[ match(colnames(counts), anno$Foreign.ID),] # check after resorting all(anno$Foreign.ID == colnames(counts)) #------------------------------------------------- #---------Filter genes #------------------------------------------------- # For now, we are not interested in making comparisons with the total RNA samples, so we exclude those before doing the normalization (and gene filtering) and running DESeq anno <- anno[ anno$group == "nascent", ] counts <- counts[,match(anno$Foreign.ID, colnames(counts))] # do simple cpm computation without any library factors, to exclude lowly expressed genes cpm_counts <- edgeR::cpm(counts, normalized.lib.sizes = FALSE) # what does the read count distribution look like? # compute median per gene and plot it as histogram median_genecounts <- apply(cpm_counts, 1, FUN=median) ggplot()+ geom_histogram(aes(log(median_genecounts+1)), bins=30) #Filter to keep genes that have a cpm of at least 0.2 in at least 1 samples print("Filtering out lowly expressed genes") keepRows <- rowSums(cpm_counts >=0.2) >= 1 table(keepRows) median_genecounts <- median_genecounts[keepRows] ggplot()+ geom_histogram(aes(log(median_genecounts+1)), bins=30) # use this to filter the gene counts matrix counts <- counts[keepRows,] #------------------------------------------------- #---------Normalize counts #------------------------------------------------- # Create DESeq2Dataset object dds <- DESeqDataSetFromMatrix(countData = counts, colData = anno, design= ~ treat) # relevel to set Vehicle treated sample as reference dds$treat <- relevel(dds$treat, ref = "Vehicle") # this would also be done automatically when calling the DESeq() function dds <- estimateSizeFactors(dds) # NOTE: DESeq2 doesn’t actually use normalized counts, rather it uses the raw counts and models the normalization inside the Generalized Linear Model (GLM). # These normalized counts will be useful for downstream visualization of results, but cannot be used as input to DESeq2 or any other tools that perform differential expression analysis which use the negative binomial model. normalized_counts <- counts(dds, normalized=TRUE) write.table(normalized_counts, paste0(opt$outdir,"/DESeq_normalized_counts_nototal.tsv"), col.names = TRUE, row.names = TRUE, sep="\t", quote=FALSE) saveRDS(normalized_counts, paste0(opt$outdir,"/DESeq_normalized_counts_nototal.rds")) write.table(anno, paste0(opt$outdir,"/anno_nototal.tsv"), col.names = TRUE, row.names = FALSE, sep="\t", quote=FALSE) #------------------------------------------------- #---------Run DESeq2 #------------------------------------------------- print("Running DESeq") dds <- DESeq(dds) res <- results(dds) # based on the computed coefficients, we now compute a contrast of LPS vs LPS+Dex res_DexLPSvLPS <- results(dds, contrast=c("treat","LPS_Dex","LPS")) res_LPSvVeh <- results(dds, contrast=c("treat","LPS","Vehicle")) res_DexLPSvVeh <- results(dds, contrast=c("treat","LPS_Dex","Vehicle")) saveRDS(res_DexLPSvLPS, paste0(opt$outdir,file = "/contrast_DexVSDexLPS.rds")) saveRDS(res_LPSvVeh, paste0(opt$outdir,file = "/contrast_LPSvVeh.rds")) saveRDS(res_DexLPSvVeh, paste0(opt$outdir,file = "/contrast_DexLPSvVeh.rds")) #------------------------------------------------- #---------Annotate genes #------------------------------------------------- print("Annotating genes") f_genekey <- paste0(opt$outdir,"/geneKey_biomart_mm_k",opt$biomart_ensembl_version,".txt") if (!file.exists(f_genekey)){ ensembl_mm <- biomaRt::useEnsembl(biomart = 'ensembl', dataset="mmusculus_gene_ensembl", version = as.numeric(opt$biomart_ensembl_version)) # retrieve the geneKey to map ensembl IDs to mgi_symbols geneKey<- biomaRt::getBM(mart=ensembl_mm, attributes=c("ensembl_gene_id","gene_biotype","mgi_symbol")) write.table(geneKey, f_genekey, sep="\t") } else { geneKey <- read.table(f_genekey, header=TRUE, sep="\t") } print("Done getting genekey") #merge gene annotations to results table res_DexLPSvLPS_ext <- merge(as.data.frame(res_DexLPSvLPS), geneKey, by.x="row.names", by.y="ensembl_gene_id", all.x=TRUE) res_LPSvVeh_ext <- merge(as.data.frame(res_LPSvVeh), geneKey, by.x="row.names", by.y="ensembl_gene_id", all.x=TRUE) res_DexLPSvVeh_ext <- merge(as.data.frame(res_DexLPSvVeh), geneKey, by.x="row.names", by.y="ensembl_gene_id", all.x=TRUE) #------------------------------------------------- #---------Export contrasts #------------------------------------------------- print("Exporting the contrasts") write.table(res_DexLPSvLPS_ext,paste0(opt$outdir,"/res_DexLPSvLPS_ext.tsv"), col.names = TRUE, row.names = FALSE, sep="\t", quote=FALSE) write.table(res_LPSvVeh_ext,paste0(opt$outdir,"/res_LPSvVeh_ext.tsv"), col.names = TRUE, row.names = FALSE, sep="\t", quote=FALSE) write.table(res_DexLPSvVeh_ext,paste0(opt$outdir,"/res_DexLPSvVeh_ext.tsv"), col.names = TRUE, row.names = FALSE, sep="\t", quote=FALSE) |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("-n", "--norm"), type="character", help="Path to normalized counts"), make_option(c("--contrast"), type="character", help="Path to annotated tsv file of DeSeq2 contrast"), make_option(c("-a", "--annotation"), type="character", help="Path to metadata with annotation"), make_option(c( "--log2fcthresh"), type="numeric", help="Log2FC threshold used in addition to adj.pval to define significant genes"), make_option(c("-o", "--outdir"), type="character", help="Path of output directory") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(ComplexHeatmap, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(topGO, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=6, family = "ArialMT", colour="black"), title=element_text(size=8, family="ArialMT", colour="black"), panel.grid.major = element_line(colour="grey", size=0.2), panel.grid.minor = element_blank(), axis.text = element_text(size=6, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(6, 'points'), #change legend key size legend.key.height = unit(6, 'points'), #change legend key height legend.key.width = unit(6, 'points'), #change legend key width legend.text = element_text(size=6, family="ArialMT", colour="black") ) dir.create(opt$outdir) res_DexLPSvLPS_ext <- read.delim( opt$contrast ) #------------------------------------------------- #---------Check for enrichements #------------------------------------------------- # Note: GenTable returns the scores of topGOresult as character which is a problem, since some pvalues are simply returned as "1e-30" and turn into NA when coercing back to numeric # that's why why made our own custom version of it, where we set the eps cutoff for these small values to FALSE source("./src/scripts/mytopGOGenTable.R") maketopGO <- function(ont, all_genes) { topgo <- new("topGOdata", description = "Test", ontology = ont, allGenes = all_genes, nodeSize = 20, annot = annFUN.org, mapping="org.Mm.eg.db", ID="ensembl") # After initializing the topGO object, we can perform the significance tests with a number of differene methods resultFisher <- runTest(topgo, algorithm = "classic", statistic = "fisher") resultFisher.parentchild <- runTest(topgo, algorithm = "parentchild", statistic = "fisher") showSigOfNodes(topgo, score(resultFisher.parentchild), firstSigNodes = 10, useInfo ='all') # initialize empty list for results we want to return later results = list() #print("Top 50 nodes sorted by Fisher parentchild") res_gentable <- mytopGOGenTable(topgo, classicFisher = resultFisher, parentchildFisher = resultFisher.parentchild, #classicKS = resultKS, elimKS = resultKS.elim, orderBy = "parentchildFisher", ranksOf = "classicFisher", topNodes = 50) results[["gentable"]] <- res_gentable # also retrieve and return the gene mapping of those top significant categories # get the GO IDs top_GOs <- res_gentable %>% mutate(parentchildFisher = as.numeric(parentchildFisher)) %>% mutate(classicFisher = as.numeric(classicFisher)) %>% filter(parentchildFisher<0.05) %>% top_n(50, wt=-parentchildFisher) %>% pull(GO.ID) # save the genes within them and their scores to a list for (GO in top_GOs){ df <- data.frame(scores=unlist(scoresInTerm(topgo,GO)), genes=unlist(genesInTerm(topgo,GO))) DEgenes <- df %>% filter(scores==2) GOname <- res_gentable %>% filter(GO.ID == GO) %>% pull(Term) GOname <- paste(GO, ":", GOname) results[[GOname]] <- DEgenes$genes } return(results) } # We create a binary vector indicating which of the investigates genes came up as significant all_sig <- as.integer(res_DexLPSvLPS_ext$padj < 0.05) names(all_sig) <- res_DexLPSvLPS_ext$Row.names up_sig <- as.integer( (res_DexLPSvLPS_ext$padj < 0.05) & (res_DexLPSvLPS_ext$log2FoldChange > opt$log2fcthresh) ) names(up_sig) <- res_DexLPSvLPS_ext$Row.names down_sig <- as.integer( (res_DexLPSvLPS_ext$padj < 0.05) & (res_DexLPSvLPS_ext$log2FoldChange < (-opt$log2fcthresh)) ) names(down_sig) <- res_DexLPSvLPS_ext$Row.names for (cat in c("BP", "MF")){ # iterate over GP categories of interest for (set in c("all","down","up")){ #iterate over our genesets print(cat) print(set) if (!file.exists(paste0("results/current/rnaseq_4sU/figures/topGO_enrichment_",cat,"_",set,".rds")) ){ # open graphics device, that the outgenerated plot will get saved to when calling topGO png(paste0(opt$outdir,"/topGO_enrichment_",cat,"_",set,"_network.png")) assign(paste0(cat,"_",set), maketopGO(cat, factor(get(paste0(set,"_sig"))) ) ) dev.off() # save the returned object as rds, so it doesn't have to get rerun next time saveRDS(get(paste0(cat,"_",set)), file=paste0("results/current/rnaseq_4sU/figures/topGO_enrichment_",cat,"_",set,".rds")) } else { assign(paste0(cat,"_",set), readRDS(file=paste0("results/current/rnaseq_4sU/figures/topGO_enrichment_",cat,"_",set,".rds")) ) } } } # We start by looking into the "BP"(=Biological Process) ontology # pdf(paste0(opt$outdir,"/topGO_enrichment_BP_all_network.pdf"), width=6, height=7, useDingbats = F, pointsize=5) # BP_all <- maketopGO("BP", factor(all_sig)) # dev.off() # pdf(paste0(opt$outdir,"/topGO_enrichment_BP_down_network.pdf"), width=6, height=7, useDingbats = F, pointsize=5) # BP_down <- maketopGO("BP", factor(down_sig)) # dev.off() # pdf(paste0(opt$outdir,"/topGO_enrichment_BP_up_network.pdf"), width=6, height=7, useDingbats = F, pointsize=5) # BP_up <- maketopGO("BP", factor(up_sig)) # dev.off() # And now for the molecular function # pdf(paste0(opt$outdir,"/topGO_enrichment_MF_all_network.pdf"), width=6, height=7, useDingbats = F, pointsize=5) # MF_all <- maketopGO("MF", factor(all_sig)) # dev.off() # pdf(paste0(opt$outdir,"/topGO_enrichment_MF_down_network.pdf"), width=6, height=7, useDingbats = F, pointsize=5) # MF_down <- maketopGO("MF", factor(down_sig)) # dev.off() # pdf(paste0(opt$outdir,"/topGO_enrichment_MF_up_network.pdf"), width=6, height=7, useDingbats = F, pointsize=5) # MF_up <- maketopGO("MF", factor(up_sig)) # dev.off() for (res in c("BP_all","BP_up","BP_down","MF_all","MF_up","MF_down")){ gg <- get(res)[[1]] %>% mutate(parentchildFisher=as.numeric(parentchildFisher))%>% top_n(30, wt=-parentchildFisher) %>% mutate(hitsperc=Significant*100/Annotated) %>% ggplot(aes(x=-log10(parentchildFisher), y=reorder(Term,-parentchildFisher), colour=hitsperc, size=Annotated)) + geom_point() + scale_colour_gradient(low = "skyblue3", high = "purple2", guide="colourbar",limits=c(0,100))+ guides(colour = guide_colourbar(barheight = 7))+ geom_vline(xintercept=-log10(0.05), linetype="dashed")+ expand_limits(x=0) + labs(x="-log10(pvalue)", y="GO term", colour="Hits (%)", size="Termsize") ggsave(paste0(opt$outdir,"/topGO_enrichment",res,".png"), gg) assign(paste0("gg_enrichment_",res),gg) } #------------------------------------------------- # #------------------------------------------------- anno <- read.delim(opt$annotation) #------------------------------------------------- #---------library normalization #------------------------------------------------- normalized_counts <- read.delim(opt$norm, header = TRUE) counts_long <- as.data.frame(normalized_counts) %>% tidyr::pivot_longer(everything(), names_to="sample", values_to = "counts") counts_boxplot <- ggplot(data=counts_long)+ geom_boxplot(aes(x=sample, y=log(counts)))+ theme(axis.text.x = element_text(angle=90)) counts_boxplot ggsave(paste0(opt$outdir,"/counts_boxplot.png"), counts_boxplot) #------------------------------------------------- #---------PCA #------------------------------------------------- # Run PCA on log of normalized counts #PCA_all = prcomp(log2(normalized_counts+1) %>% t(), # center=TRUE, # scale=FALSE) #center to change mean to 0, scale to change SD to 1 transposed_variables <- as.data.frame(t(log2(normalized_counts+1))) %>% mutate(treat=factor(anno$treat, levels=c("Vehicle","LPS","LPS_Dex")) ) %>% dplyr::select("treat",everything()) PCA_all <- FactoMineR::PCA(X = transposed_variables, # transposed so that the variables (=genes) are columns scale.unit = TRUE, # whether we scale the variance or not (centering gets done automatically) ncp = 5, # Number of PCs quali.sup = 1, # Position of the qualitative variable graph = FALSE) # No graphic outputs (lib. factoextra) gg_elbow <- factoextra::fviz_eig(PCA_all, addlabels = FALSE, ncp=10, main=NULL, ylab="% var", ggtheme = theme(plot.title = element_blank()) ) gg_pca1 <- factoextra::fviz_pca_ind(PCA_all, #col.ind="cos2", axes=c(1,2), pointsize=4, habillage = PCA_all$call$X$treat, legend.title = "", title="", #addEllipses = TRUE, #too few points for ellipse label=FALSE, palette = c("#CD8500","#8DA0CB","#66C2A5"), #c("veh" = "#CD8500", "lps" = "#8DA0CB","lps_dex" = "#66C2A5")) invisible="quali") # in order to disable displaing centroid gg_pca2 <- factoextra::fviz_pca_ind(PCA_all, #col.ind="cos2", axes=c(3,4), pointsize=4, habillage = PCA_all$call$X$treat, legend.title = "", #addEllipses = TRUE, #too few points for ellipse label=FALSE, palette = c("#CD8500","#8DA0CB","#66C2A5"), #c("veh" = "#CD8500", "lps" = "#8DA0CB","lps_dex" = "#66C2A5")) invisible="quali") # in order to disable displaing centroid gg_pca <- ggpubr::ggarrange(gg_pca1,gg_pca2, common.legend=TRUE) gg_pca ggsave(paste0(opt$outdir,"/PCA.png"),gg_pca) #------------------------------------------------- #---------Volcano plots #------------------------------------------------- res_DexLPSvLPS_ext <- res_DexLPSvLPS_ext %>% mutate(change=case_when(padj<0.05 & log2FoldChange > opt$log2fcthresh ~ "sig up", padj<0.05 & log2FoldChange < (-opt$log2fcthresh) ~ "sig down", TRUE ~ "no change")) table(res_DexLPSvLPS_ext$change) gg_volcano <- res_DexLPSvLPS_ext %>% ggplot()+ geom_point(aes(x=log2FoldChange, y=-log10(padj), colour=change))+ coord_cartesian(ylim=c(-5,120),xlim=c(-8,8))+ scale_color_manual(name="", values = c("no change"="#000000","sig down"="#8DA0CB","sig up"="#66C2A5"))+ ggrepel::geom_text_repel(data=dplyr::filter(res_DexLPSvLPS_ext,-log10(padj) > 28 | abs(log2FoldChange) > 6), aes(x=log2FoldChange, y=-log10(padj), label=mgi_symbol, colour=change), size=2, min.segment.length = 0, max.overlaps = 100, box.padding = 0.1, show.legend = FALSE)+ labs(title="DexLPS vs LPS")+ theme(legend.position = c(0.15, 0.8)) gg_volcano ggsave(paste0(opt$outdir,"/volcano_plot_DexLPSvLPS.png"), gg_volcano) #------------------------------------------------- #---------Heatmap, ALL genes #------------------------------------------------- # For the visualizations in heatmaps, we use the normalized counts (counts normalized by DESeq's sizefactors) and add a pseudocount (+1) before taking the log. # We then z-scale those log-transformed counts row-wise (for each gene). LOG.all_zn <- t(apply(log(normalized_counts+1), 1, function(x) (x-mean(x))/sd(x) ) ) #z-scale normalized values by row (by gene) #create annotation labels for the heatmap ha.tmp <- HeatmapAnnotation(df = anno %>% dplyr::select("treat"), col = list(treat = c("Vehicle" = "#CD8500", "LPS" = "#8DA0CB", "LPS_Dex" = "#66C2A5")),#red,green, blue show_annotation_name = c(bar = FALSE), show_legend = c(treat=TRUE), annotation_legend_param = list( labels_gp = gpar(fontsize=6), title_gp = gpar(fontsize=6) ) #subset=setNames(RColorBrewer::brewer.pal(6,"Set1") ,levels(met.all$subset)) #annotation_legend_param=list( ) #instead of manually assigning each value a color, we can just assign our factor levels the colours of a brewer pallette ) heatmap_allgenes <- Heatmap( LOG.all_zn, top_annotation = ha.tmp, column_title="All genes, all samples", column_title_gp = gpar(fontsize = 6, fontface = "bold"), name="Row Z-Score", #Title on top of legend clustering_distance_rows = "euclidean", clustering_method_rows = "complete", show_row_dend = FALSE, show_row_names = FALSE, show_column_names = FALSE, column_dend_height = unit(1, "cm"), row_names_gp = gpar(fontsize = 6), heatmap_legend_param = list( title_position = "leftcenter-rot", legend_direction="vertical", labels_gp = gpar(fontsize=6), title_gp = gpar(fontsize=6), legend_height = unit(2, "cm"), at = c(-2, 0, 2), labels = c("-2", "0", "2") ) ) pdf(paste0(opt$outdir,"/heatmap_allgenes.pdf"), width=6, height=7, useDingbats = F, pointsize=5) draw(heatmap_allgenes, heatmap_legend_side="right", merge_legend = TRUE) dev.off() gg_heatmap_allgenes <- grid.grabExpr(draw(heatmap_allgenes, heatmap_legend_side="right", merge_legend = TRUE)) gg_heatmap_allgenes #------------------------------------------------- #---------Heatmap, SIGgenes #------------------------------------------------- #remove those entries in results that have NA as padj and filter for the significant ones res_DexLPSvLPS_sig <- res_DexLPSvLPS_ext %>% filter(!is.na(padj)) %>% filter(padj<0.05) %>% filter(abs(log2FoldChange) > opt$log2fcthresh) heatmap_siggenes <- Heatmap( LOG.all_zn[res_DexLPSvLPS_sig$Row.names,] , top_annotation = ha.tmp, column_title="DexLPSvsLPS DE genes", column_title_gp = gpar(fontsize = 6, fontface = "bold"), name="Row Z-Score", #Title on top of legend clustering_distance_rows = "euclidean", clustering_method_rows = "complete", show_row_dend = FALSE, show_row_names = FALSE, show_column_names = FALSE, column_dend_height = unit(1, "cm"), row_names_gp = gpar(fontsize = 6), heatmap_legend_param = list( title_position = "leftcenter-rot", legend_direction="vertical", labels_gp = gpar(fontsize=6), title_gp = gpar(fontsize=6), legend_height = unit(2, "cm"), at = c(-2, 0, 2), labels = c("-2", "0", "2") ) ) pdf(paste0(opt$outdir,"/heatmap_siggenes_DexLPSvsLPS.pdf"), width=6, height=7, useDingbats = F, pointsize=5) draw(heatmap_siggenes, heatmap_legend_side="bottom") dev.off() gg_heatmap_siggenes <- grid.grabExpr(draw(heatmap_siggenes, heatmap_legend_side="right", merge_legend = TRUE)) #--------------------------------------------------------------- #---------Heatmap for DE genes within GO category of interest #--------------------------------------------------------------- # "GO:0060089 : molecular transducer activity" # "GO:0001067 : transcription regulatory region nucleic ..." mapping <- res_DexLPSvLPS_ext %>% filter(padj<0.05 & abs(log2FoldChange) > opt$log2fcthresh) %>% filter(Row.names %in% MF_all[["GO:0060089 : molecular transducer activity"]]) %>% dplyr::select(c(Row.names,mgi_symbol)) heatmap_GO <- Heatmap( LOG.all_zn[mapping$Row.names,], top_annotation = ha.tmp, column_title="Sig genes in GO:0060089 : \n molecular transducer activity", column_title_gp = gpar(fontsize = 6, fontface = "bold"), name="Row Z-Score", #Title on top of legend clustering_distance_rows = "euclidean", clustering_method_rows = "complete", row_labels = mapping$mgi_symbol, show_row_dend = FALSE, show_row_names = TRUE, show_column_names = FALSE, column_dend_height = unit(1, "cm"), row_names_gp = gpar(fontsize = 5), heatmap_legend_param = list( title_position = "leftcenter-rot", legend_direction="vertical", labels_gp = gpar(fontsize=6), title_gp = gpar(fontsize=6), legend_height = unit(2, "cm"), at = c(-2, 0, 2), labels = c("-2", "0", "2") ) ) heatmap_GO pdf(paste0(opt$outdir,"/heatmap_siggenesGO0_DexLPSvsLPS.pdf"), width=6, height=7, useDingbats = F, pointsize=5) draw(heatmap_GO, heatmap_legend_side="bottom") dev.off() gg_heatmap_GO <- grid.grabExpr(draw(heatmap_GO, heatmap_legend_side="right", merge_legend = TRUE)) #------------------------------------------------------------------------------------ #------ include the GO terms #------------------------------------------------------------------------------------ GO_network <- png::readPNG("results/current/rnaseq_4sU/figures/topGO_enrichment_MF_all_network.png") gg_GO_network <- ggplot() + ggpubr::background_image(GO_network) + # so it doesn't get squished coord_fixed()+ # This ensures that the image leaves some space at the edges theme(plot.margin = margin(t=0, l=0, r=0, b=0, unit = "cm"), axis.line = element_blank()) #--------------------------------------------------------------- #---------Pull the figures together #--------------------------------------------------------------- #lay <- rbind(c(1,2,2), # c(3,4,7), # c(3,4,7), # c(5,5,6), # c(5,5,6)) #m1 <- gridExtra::grid.arrange(gg_elbow, gg_pca2, # gg_heatmap_allgenes, gg_volcano, # gg_enrichment_MF_all, gg_heatmap_GO0016772, # gg_GO_network, # layout_matrix=lay) #ggsave("results/current/Figure_2.png",m1, width = 20, height = 30, units = "cm") gg_r1 <- ggpubr::ggarrange(gg_elbow, gg_pca1, labels = c("A", NA), ncol = 2, nrow = 1, widths=c(1.4,1)) gg_r2c1 <- ggpubr::ggarrange(gg_volcano, gg_enrichment_MF_all, gg_GO_network, labels = c("B", "D","F"), ncol = 1, nrow = 3, heights = c(1,1,0.5)) gg_r2c2 <- ggpubr::ggarrange(gg_heatmap_siggenes, gg_heatmap_GO, labels = c("C", "E"), ncol = 1, nrow=2, heights=c(1,1)) gg_r2 <- ggpubr::ggarrange(gg_r2c1, gg_r2c2, labels = c(NA, NA), ncol = 2, nrow=1, widths=c(1,1)) full_panel <- ggpubr::ggarrange(gg_r1, gg_r2, nrow=2, heights=c(0.4,2)) full_panel ggsave("results/current/Figures/Figure_rnaseq.png", full_panel, width=190, height=250, units="mm", bg="white") ggsave("results/current/Figures/Figure_rnaseq.pdf", full_panel, width=190, height=250, units="mm", bg="white") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--summitAnno"), type="character", help="Path to dataframe object of GR summits annotated to closest gene"), make_option(c("--chipseq_peaks"), type="character", help="Path to IDR ChIPseq peaks"), make_option(c("--chipseq_summits"), type="character", help="Path to summit file of IDR peaks"), make_option(c("--sm_summitranges"), type="character", help="Path to rds file of genomation score matrix around peak summits"), make_option(c("--nr3c1fullsitematches"), type="character", help="Path to homer hits of nr3c1 fullsite motif"), make_option(c("--nr3c1halfsitematches"), type="character", help="Path to homer hits of nr3c1 halfsite motif"), make_option(c( "--streme"), type="character", help="Path to XML streme output file of enrichment around peak summits")) opt <- parse_args(OptionParser(option_list=option_list)) # set output for logfile to retrieve stats for plot later sink(file="results/current/figure_chipseq.out") suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(grid, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(memes, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(universalmotif, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(rtracklayer, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ChIPseeker, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(genomation, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=12, family = "ArialMT", colour="black"), title=element_text(size=16, family="ArialMT", colour="black"), panel.grid.major = element_line(colour="grey", size=0.2), panel.grid.minor = element_blank(), axis.text = element_text(size=12, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(12, 'points'), #change legend key size legend.key.height = unit(12, 'points'), #change legend key height legend.key.width = unit(12, 'points'), #change legend key width legend.text = element_text(size=8, family="ArialMT", colour="black") ) #---------------------------------------------- #------ load input data #---------------------------------------------- chipseq <- rtracklayer::import.bed(opt$chipseq_peaks) chipseq <- GenomeInfoDb::keepStandardChromosomes(chipseq, pruning.mode = "tidy") names(chipseq) <- c(1:length(chipseq)) print(paste("We find a total of ", length(chipseq),"ChIPseq peaks")) chipseq_summits <- read.table(opt$chipseq_summits) chipseq_summits <- GRanges(seqnames = chipseq_summits[,c("V1")], ranges = IRanges(start=chipseq_summits[,c("V2")], end=chipseq_summits[,c("V3")]-1)) # to make up for 0 vs 1 encoding chipseq_summits$id <- c(1:length(chipseq_summits)) chipseq_summitranges <- chipseq_summits %>% plyranges::anchor_center() %>% plyranges::mutate(width = 1000) sm_summitranges <- readRDS(opt$sm_summitranges) #---------------------------------------------- #------ peak width #---------------------------------------------- xbreaks <- c(200,500,1000,3000) gg_peakwidth <- ggplot(as.data.frame(chipseq))+ geom_histogram(aes(log10(width)), bins=50, colour="black")+ scale_x_continuous("width (bp)", breaks=log10(xbreaks), labels=xbreaks, limits=log10(c(200,4000)) ) gg_peakwidth print(paste("The peak width has a mean of", format(mean(width(chipseq)), digits = 6), "and a median of", format(median(width(chipseq)),digits = 6))) #---------------------------------------------- #------ peak wrt reference #---------------------------------------------- summitAnno <- readRDS(opt$summitAnno) ## 1. where do the peaks lie #------------------------------- annostat <- as.data.frame(summitAnno@annoStat) gg_annopie <- ggplot(annostat, aes(x="", y=Frequency, fill=Feature))+ geom_bar(stat="identity", width=1, color="white") + geom_text(aes(label = paste(round(Frequency, 1), "%"), x = 1.5), position = position_stack(vjust = 0.5), size=2) + coord_polar("y", start=0)+ #scale_fill_brewer(palette="Set3")+ scale_fill_manual(values= c(rev(RColorBrewer::brewer.pal(9,"YlGnBu")),"white", "darkseagreen"))+ theme_void()+ theme( text=element_text(size=6, family = "ArialMT", colour="black"), legend.key.height = unit(6, 'points'), #change legend key height legend.key.width = unit(6, 'points'), #change legend key width legend.text = element_text(size=6, family="ArialMT", colour="black") ) gg_annopie print(paste("Frequency of peaks within promoters <3kb:", format(sum(annostat[1:3,"Frequency"]),digits=4))) print(paste("Frequency of peaks in introns:", format(sum(annostat[8:9,"Frequency"]),digits=4))) print(paste("Frequency of peaks in distal intergenic regions:", format(sum(annostat[11,"Frequency"]),digits=4))) ## 2. how far are they from closest TSS #------------------------------- xbreaks <- c(-100, -75, -50, -25, 0, 25, 50, 75, 100) gg_distexpr <- as.data.frame(summitAnno) %>% ggplot( aes(x=distanceToTSS)) + geom_histogram(binwidth = 3000) + coord_cartesian( # ylim=c(0,2000), xlim=c(-100*10^3, 100*10^3) )+ geom_segment(aes(x = 30*1000, y = 0, xend = 30*1000, yend = 3000), colour="black", linetype=2)+ geom_segment(aes(x = -30*1000, y = 0, xend = -30*1000, yend = 3000), colour="black", linetype=2)+ scale_x_continuous( breaks = xbreaks*1000, labels = paste(xbreaks, "kb") )+ labs(title="", y="Counts")+ theme(axis.text.x = element_text(angle=45, vjust=1, hjust=1)) gg_distexpr print(paste("Out of all our peaks, a total of", as.data.frame(summitAnno) %>% filter(abs(distanceToTSS) < 30000) %>% nrow(), "where within 30000kb of a TSS")) print(paste("This corresponds to", format(as.data.frame(summitAnno) %>% filter(abs(distanceToTSS) < 30000) %>% nrow() / length(chipseq) *100, digits=4), "%")) print(paste("They annotated to ", as.data.frame(summitAnno) %>% filter(abs(distanceToTSS) < 30000) %>% pull(SYMBOL) %>% unique() %>% length(), "unique gene symbols")) #---------------------------------------------- #------ load motif query and match #---------------------------------------------- load_sitematches <- function(bedpath){ sitematches <- rtracklayer::import.bed(bedpath) # remove weird repetition of seqname seqlevels(sitematches) <- gsub(" .*$","",seqlevels(sitematches)) sitematches <- GenomeInfoDb::keepStandardChromosomes(sitematches, pruning.mode = "tidy") names(sitematches) <- c(1:length(sitematches)) return(sitematches) } nr3c1fullsitematches <- load_sitematches(opt$nr3c1fullsitematches) nr3c1halfsitematches <- load_sitematches(opt$nr3c1halfsitematches) print(paste("Genome-wide, we found", length(nr3c1fullsitematches), "matches for the fullsite and", length(nr3c1halfsitematches), "matches for the halfsite")) #---------------------------------------------- #------ # counts for overlap with nr3c1 motif #---------------------------------------------- # crashes if including halfsites #overlap_wfullsite <- ChIPpeakAnno::findOverlapsOfPeaks(chipseq, # nr3c1fullsitematches, # nr3c1halfsitematches, # minoverlap=1) #gg_venn <- grid.grabExpr( # ChIPpeakAnno::makeVennDiagram(overlap_wfullsite, # fill=c("#669933", "#ff9900", "#c01311"), # circle fill color (green, orange) # col=c("#669933", "#ff9900", "#c01311"), #circle border color # cat.col=c("#669933", "#ff9900", "#c01311"), # method = NULL, # cex = 0.6, # connectedPeaks = "keepFirstListConsistent"), # vp = viewport(w = .6, h = 1.0) #) chipseq_nr3c1fullsitematches_counts <- plyranges::count_overlaps(chipseq, nr3c1fullsitematches) %>% as.data.frame() %>% magrittr::set_colnames(c("sitecounts")) %>% count(sitecounts) %>% mutate(freq = n / sum(n), site = "full") print(paste("The number of peaks which have at least 1 fullsite match:", sum(chipseq_nr3c1fullsitematches_counts$n) - chipseq_nr3c1fullsitematches_counts$n[1] )) print(paste("This corresponds to", sum(chipseq_nr3c1fullsitematches_counts$freq) - chipseq_nr3c1fullsitematches_counts$freq[1], "%" )) chipseq_nr3c1halfsitematches_counts <- plyranges::count_overlaps(chipseq, nr3c1halfsitematches)%>% as.data.frame() %>% magrittr::set_colnames(c("sitecounts")) %>% count(sitecounts) %>% mutate(freq = n / sum(n), site="half") print(paste("The number of peaks which have at least 1 halfsite match:", sum(chipseq_nr3c1halfsitematches_counts$n) - chipseq_nr3c1halfsitematches_counts$n[1] )) print(paste("This corresponds to", sum(chipseq_nr3c1halfsitematches_counts$freq) - chipseq_nr3c1halfsitematches_counts$freq[1], "%" )) rm(nr3c1fullsitematches) rm(nr3c1halfsitematches) chipseq_nr3c1sitematches_counts <- rbind(chipseq_nr3c1fullsitematches_counts,chipseq_nr3c1halfsitematches_counts) chipseq_nr3c1sitematches_counts <- chipseq_nr3c1sitematches_counts %>% mutate(sitecounts_fac = case_when(sitecounts<=15 ~ as.character(sitecounts), sitecounts>15 ~ ">15"), # aggregate it towards to top end sitecounts_fac = factor(as.character(sitecounts_fac)) ) # set order of factor levels my_levels <- levels(chipseq_nr3c1sitematches_counts$sitecounts_fac)[gtools::mixedorder(levels(chipseq_nr3c1sitematches_counts$sitecounts_fac))] chipseq_nr3c1sitematches_counts <- chipseq_nr3c1sitematches_counts %>% mutate(sitecounts_fac = factor(sitecounts_fac, levels=my_levels)) %>% group_by(site, sitecounts_fac) %>% summarize(freq=sum(freq), total=sum(n)) gg_barplotmotifhits <- ggplot(data=chipseq_nr3c1sitematches_counts, aes(x=sitecounts_fac, y= freq*100, group=site))+ geom_bar(aes(colour=site, fill=site), stat="identity", alpha=0.7, position = position_dodge2(width=0.4, padding=0.1, preserve = "single") )+ scale_fill_manual("", labels = c("NR3C1 fullsite", "NR3C1 halfsite"), breaks = c("full", "half"), values = c("black", "darkgrey")) + scale_colour_manual("", labels = c("NR3C1 fullsite", "NR3C1 halfsite"), breaks = c("full", "half"), values = c("black", "darkgrey"))+ labs(x="# of motifmatches", y="% of ChIPseq peaks")+ theme(legend.position = c(0.7, 0.8), axis.text.x = element_text(angle=45, vjust=1, hjust=1)) gg_barplotmotifhits # can we use the scorematrix function from genomation to get a distribution of reads wrt handpicked motifs #------------------------------------------------------------------------------------ #------ memes #------------------------------------------------------------------------------------ # conda activate py_3 # perlbrew use perl-5.34.0 # nohup Rscript memes_runanalyses.R & (from within the script directory) #------ STREME #----------------- streme_results <- memes::importStremeXML( opt$streme ) streme_results <- streme_results %>% mutate(name = paste(consensus, pval)) #DT::datatable(streme_results %>% dplyr::relocate("eval",.after="consensus"), # extensions = c('Buttons'), # width=1080, # options = list(dom = 'Bfrtip', buttons = c('csv', 'excel'), scrollX = TRUE), # caption = htmltools::tags$caption(style = 'caption-side: bottom; text-align: center;', htmltools::em('Motif enrichment - Discriminative analysis')) #) gg_streme <- streme_results[1:5,] %>% to_list() %>% view_motifs(names.pos = "top", tryRC = FALSE # we don't care about maximizing based on alignment score, just wanna display the discovered motifs ) gg_streme # why are the lists with pos distr not the same length? Let's pad them padna <- function(myvector, outlength){ # check length of vector and how was it's off from desired length tofill <- outlength - length(myvector) if (tofill<0) { stop("Desired length is longer than current length") } else if(tofill > 1) { # check if it's even if((tofill %% 2) == 0) { outvector = c(rep(NA,tofill/2), myvector,rep(NA,tofill/2)) } else { outvector = c(rep(NA,floor(tofill/2)), myvector,rep(NA,ceiling(tofill/2))) } # if uneven, put -1 before and +1 after return(outvector) } } pos_distr <- streme_results$site_distr %>% stringr::str_trim(side = c("both")) %>% stringr::str_split(pattern=" ") pos_distr <- lapply(pos_distr, function(x) padna(x,outlength=101)) pos_distr <- as.data.frame(do.call(rbind, pos_distr)) colnames(pos_distr) <- 1:ncol(pos_distr) pos_distr_top_long <- pos_distr %>% mutate(altname = streme_results$altname) %>% filter(altname %in% paste0("STREME-",c(1,2,3,4,5)) ) %>% relocate(altname) %>% tidyr::pivot_longer(cols=1:ncol(pos_distr)+1, names_to="position", values_to = "signal") %>% mutate(position=as.numeric(position), signal=as.numeric(signal)) gg_pos_distr <- ggplot(pos_distr_top_long,aes(x=position, y=signal))+ geom_point(size=0.5)+ geom_smooth()+ facet_wrap(~altname, ncol=1, scales="free_y")+ labs(x="potision",y="frequency")+ scale_x_continuous(breaks=c(1,51,101), labels=c(-50,0,50))+ theme( strip.background = element_blank(), strip.text.x = element_blank(), axis.ticks.y = element_blank(), axis.text.y = element_blank() ) gg_pos_distr #------------------------------------------------------------------------------------ #------ read distribution around peak summit ? #------------------------------------------------------------------------------------ genomation_profiledata <- genomation::plotMeta(sm_summitranges) # make sure the xcoords match the values specified in the summitranges (or the xaxis labels will be wrong) gg_chipseq_genomationprofileplot <- ggplot()+ geom_line(aes(x=seq(1:length(genomation_profiledata)),y=genomation_profiledata))+ scale_x_continuous("bases", breaks=c(0,250,500,750,1000), labels=c(-500,-250,0,250,500))+ labs(x="bases",y="average score")+ theme(plot.margin = margin(1,2,0,0, "cm")) sm_scaled = genomation::scaleScoreMatrix(sm_summitranges) gg_chipseq_genomationheatmap <- grid.grabExpr( genomation::heatMatrix(sm_scaled, xcoords = c(-500, 500)) ) #------------------------------------------------------------------------------------ #------ plot aggregated figure #------------------------------------------------------------------------------------ gg_c1 <- ggpubr::ggarrange(gg_barplotmotifhits,gg_peakwidth, labels = c("A","B"), ncol = 1, nrow = 2, heights=c(1,0.8)) gg_c2 <- ggpubr::ggarrange(gg_chipseq_genomationprofileplot, gg_chipseq_genomationheatmap, labels = c("C",NA), ncol = 1, nrow = 2, heights=c(1,2)) gg_c3_r1 <- ggpubr::ggarrange(gg_streme, gg_pos_distr, labels = c("D",NA), ncol = 2, nrow = 1, widths = c(1,0.5)) gg_c3_r2 <- ggpubr::ggarrange(gg_annopie, gg_distexpr, labels = c("E", "F"), ncol = 2, nrow = 1, widths = c(1,1)) gg_c3 <- ggpubr::ggarrange(gg_c3_r1, gg_c3_r2, labels = c(NA,NA), ncol = 1, nrow = 2, heights=c(1,1)) full_panel <- ggpubr::ggarrange(gg_c1, gg_c2, gg_c3, labels = NA, ncol = 3, nrow=1, widths = c(1,1,2)) full_panel # usually 190(width) by 100(height) # scale it up, so motifs are displayed correctly ggsave(here("./results/current/Figures/Figure_chipseq.png"), full_panel, width=380, height=200, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_chipseq.pdf"), full_panel, width=380, height=200, units="mm", bg="white") sink() |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--chipseq_bam"), type="character", help="Path to BAM file with deduplicated reads from DexLPS condition"), make_option(c("--chipseq_summits"), type="character", help="Path to summit file of IDR peaks"), make_option(c("--nr3c1fullsitematches"), type="character", help="Path to homer hits of nr3c1 fullsite motif"), make_option(c("--nr3c1halfsitematches"), type="character", help="Path to homer hits of nr3c1 halfsite motif"), make_option(c("-o", "--outdir"), type="character", help="Path to output directory")) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(rtracklayer, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(genomation, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) #------ chipseq summits #---------------------------------------------- chipseq_summits <- read.table(opt$chipseq_summits) chipseq_summits <- GRanges(seqnames = chipseq_summits[,c("V1")], ranges = IRanges(start=chipseq_summits[,c("V2")], end=chipseq_summits[,c("V3")]-1)) # to make up for 0 vs 1 encoding chipseq_summits$id <- c(1:length(chipseq_summits)) chipseq_summitranges <- chipseq_summits %>% plyranges::anchor_center() %>% plyranges::mutate(width = 1000) #------ motifmatches #---------------------------------------------- nr3c1fullsitematches <- rtracklayer::import.bed(opt$nr3c1fullsitematches) # remove weird repetition of seqname seqlevels(nr3c1fullsitematches) <- gsub(" .*$","",seqlevels(nr3c1fullsitematches)) nr3c1fullsitematches <- GenomeInfoDb::keepStandardChromosomes(nr3c1fullsitematches, pruning.mode = "tidy") names(nr3c1fullsitematches) <- c(1:length(nr3c1fullsitematches)) nr3c1fullsitematches_ranges <- nr3c1fullsitematches %>% plyranges::anchor_center() %>% plyranges::mutate(width = 1000) #------ make subsets by combining chipseq summits and motifhits #---------------------------------------------- # only use the4 summitranges containing an nr3c1 motif chipseq_summitranges_inner_hits <- chipseq_summitranges %>% plyranges::join_overlap_inner(nr3c1fullsitematches) width(chipseq_summitranges_inner_hits) length(chipseq_summitranges_inner_hits) # only use the nr3c1 coordinates that fall within summitranges chipseq_summitranges_intersect_hits <- chipseq_summitranges %>% plyranges::join_overlap_intersect(nr3c1fullsitematches) width(chipseq_summitranges_intersect_hits) length(chipseq_summitranges_intersect_hits) # size it up from 14bp to 1000bp around the motifhit chipseq_summitranges_intersect_hits <- chipseq_summitranges_intersect_hits %>% plyranges::anchor_center() %>% plyranges::mutate(width = 1000) #---------------------------------------------- #------ compute score matrix #---------------------------------------------- # for each summitregion, get coverage of how many reads in bam file overlap it at each position # aggregate it across all peaks sm_summitranges <- ScoreMatrix(target = opt$chipseq_bam, windows = chipseq_summitranges, weight.col = "score") sm_nr3c1fullsitematches_ranges <- ScoreMatrix(target = opt$chipseq_bam, windows = nr3c1fullsitematches_ranges, weight.col = "score") sm_summitranges_w_nr3c1fullsitehit <- ScoreMatrix(target = opt$chipseq_bam, windows = chipseq_summitranges_inner_hits, weight.col = "score") sm_nr3c1fullsitehits_within_summitranges <- ScoreMatrix(target = opt$chipseq_bam, windows = chipseq_summitranges_intersect_hits, weight.col = "score") test_coverage <- chipseq_summitranges_intersect_hits %>% plyranges::compute_coverage() score(test_coverage) #---------------------------------------------- #------ export data #---------------------------------------------- saveRDS(sm_summitranges, paste0(opt$outdir,"sm_summitranges.rds")) saveRDS(sm_nr3c1fullsitematches_ranges, paste0(opt$outdir,"sm_nr3c1fullsitematches_ranges.rds")) saveRDS(sm_summitranges_w_nr3c1fullsitehit, paste0(opt$outdir,"sm_summitranges_w_nr3c1fullsitehit.rds")) saveRDS(sm_nr3c1fullsitehits_within_summitranges, paste0(opt$outdir,"sm_nr3c1fullsitehits_within_summitranges.rds")) |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--model_coefs_joint"), type="character", help="Path to rds file with model coefficients of joint models"), make_option(c("--model_coefs_sep"), type="character", help="Path to rds file with model coefficients of models tha include enhancers and promoters separately"), make_option(c("--auc"), type="character", help="Path to rds file with auc results of all models"), make_option(c("--motifcounts_summitregion"), type="character", help="Path to rds file with motifcounts within summitregions"), make_option(c("--raw_counts"), type="character", help="Path to rds file of raw counts within the prox based model") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggcorrplot, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(RColorBrewer, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(circlize, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggpubr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ComplexHeatmap, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=6, family = "ArialMT", colour="black"), title=element_text(size=8, family="ArialMT", colour="black"), panel.grid.major = element_line(colour="grey", size=0.2), panel.grid.minor = element_blank(), axis.text = element_text(size=6, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(6, 'points'), #change legend key size legend.key.height = unit(6, 'points'), #change legend key height legend.key.width = unit(6, 'points'), #change legend key width legend.text = element_text(size=6, family="ArialMT", colour="black") ) #---------------------------------------------- #------ load data #---------------------------------------------- model_coefs_joint <- readRDS ( opt$model_coefs_joint ) model_coefs_sep <- readRDS ( opt$model_coefs_sep ) AUC_metrics <- readRDS ( opt$auc ) AUC_metrics <- AUC_metrics %>% mutate_at(1:6,as.factor) %>% mutate(motifdata = dplyr::recode_factor(motifdata, motifcounts_abcregion = "abcregion", motifcounts_summitregion = "summitregion")) # make factor names shorter AUC_metrics <- AUC_metrics %>% mutate(names=rownames(.)) # we can't remove the rownames, since we need them for the heatmap annotation later #--------------------------------------------- #---- show correlation structure #--------------------------------------------- motifcounts_summitregion <- readRDS( opt$motifcounts_summitregion ) #motifdf <- read.table("~/projects/pipeline_ChIP-nexus/results/current/integrate_RNAseq/fimo_featurematrix/merged_matchedpeaks_bygene_30kb_slop50_thresh0.0001_withgenenames.tsv", header=TRUE) # drop metadata motifdf <- motifcounts_summitregion[,-c(1:2)] # determine most variable motifs motifdf_var <- motifdf %>% dplyr::summarise(across(where(is.numeric), var)) top30perc_var_motif <- sort(motifdf_var, decreasing = T)[1:ceiling(ncol(motifdf_var)*0.3)] %>% names() # compute and plot their correlation cor_mat_topvar <- cor(x=motifdf %>% dplyr::select(all_of(top30perc_var_motif)) ) gg_cor <- ggcorrplot(corr = cor_mat_topvar, hc.order = TRUE, # reorders variables according to their correlations outline.col = "white", colors = c("#6D9EC1", "white", "#E46726"), tl.cex=3) gg_cor ggsave(here("./results/current/Figures/Figure_motifcorrelations.png"), gg_cor, width=190, height=190, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_motifcorrelations.pdf"), gg_cor, width=190, height=190, units="mm", bg="white") #--------------------------------------------- #---- AUC barplots #--------------------------------------------- melted_AUC_metrics <- AUC_metrics %>% reshape::melt(., measure.vars=c("net_train","net_test")) # retrieve name of top 25 best performing models (to filter heatmap) top_models <- melted_AUC_metrics %>% filter(variable=="net_test") %>% arrange(desc(value)) %>% top_n(25) #(with this cutoff the prox based one gets included) top_models # overall best performance max_net_test <- melted_AUC_metrics %>% filter(variable=="net_test") %>% filter(value==max(value)) #%>% summarize(max(value)) %>% as.numeric() max_net_test #remove the proximity based model and add it in shape of a reference line instead prox_reference <- melted_AUC_metrics %>% filter(variable=="net_test") %>% filter(names=="prox") %>% pull(value) %>% as.numeric() prox_coefs <- model_coefs_joint %>% filter(prox!=0) %>% dplyr::pull(featurename) # Create the plot gg_AUC <- ggplot(melted_AUC_metrics %>% filter(variable=="net_test") %>% filter(names!="prox"))+ geom_bar(aes(x=weight, y=value, fill=condition, alpha=onlymax), colour="black", size=0.5, width=0.9, stat="identity", position=position_dodge())+ geom_hline(aes(yintercept=prox_reference, linetype="reference"), colour="purple")+ scale_linetype_manual(name="", values = c(2))+ facet_grid( cols = vars(sepPromEnh,excludepromoters), rows= vars(motifdata), labeller = labeller( motifdata=c("abcregion"="active regions", "summitregion"="GR summitregions"), excludepromoters=c("FALSE"="excl.prom: none", "onlyNONself"="excl.prom: nonself", "all"="excl.prom: all"), sepPromEnh=c("TRUE"="sep prom and enh", "FALSE"= "aggr prom and enh") ) )+ scale_fill_manual( values=c("#339966","#0066CC", "orange"), breaks=c("dexlps","lps", "DexLPS-LPS"), labels=c("DexLPS","LPS", "\u0394 DexLPS-LPS") )+ scale_alpha_manual( values=c(1,0.5), breaks=c("FALSE","TRUE"), labels=c("1-to-many", "1-to-1") )+ labs(x="Weight features during aggregation", y="AUC", alpha="mapping" )+ theme( legend.position = "bottom") gg_AUC #--------------------------------------------- #---- coefficient heatmap #--------------------------------------------- #AUC_metrics <- AUC_metrics %>% mutate_if(is.character,as.factor) make_coefficient_heatmap <- function(model_coefs){ # check what the input was to filter models accordingly later if (substitute(model_coefs) == "model_coefs_joint"){ is_separate=FALSE } else if (substitute(model_coefs) == "model_coefs_sep"){ is_separate=TRUE } else { stop("The coefficient input matrix doesn't match the expected size") } #assign any 0 coefficients an NA (so they don't get affected by scaling) model_coefs[model_coefs==0] <- NA # remove the intercept term for plotting #model_coefs <- model_coefs %>% filter(featurename!="(Intercept)") # count how many models have the motif included with a non-zero coef model_coefs_ftr <- model_coefs %>% mutate( sum = rowSums(across(where(is.numeric)), na.rm = TRUE)) %>% mutate( motifhascoef_counts = rowSums(!is.na(.)) ) %>% dplyr::select(any_of( c("featurename", "sum", "motifhascoef_counts", top_models$names))) %>% # only plot models with best performance mutate(motifhascoef_counts_topmodels = rowSums( !is.na( dplyr::select(., 3:ncol(.)) ) )) %>% mutate(., across(3:(ncol(.)-1), ~(scale(.) %>% as.vector))) %>% #scale before filtering on spec. factors slice_max(motifhascoef_counts_topmodels,n=40) # remove extra columns before passing it to heatmap model_coefs_numeric <- model_coefs_ftr %>% tibble::column_to_rownames("featurename") %>% dplyr::select(!c(sum, motifhascoef_counts, motifhascoef_counts_topmodels)) row_ha = rowAnnotation( counts = anno_barplot(model_coefs_ftr$motifhascoef_counts_topmodels) ) # take annotation from AUC_metrics #create annotation labels for the heatmap myreds <- brewer.pal(3,"OrRd") col_ha <- HeatmapAnnotation(df = AUC_metrics %>% filter(names %in% names(model_coefs_numeric)) %>% arrange(match(names,names(model_coefs_numeric))) %>% dplyr::select(c(condition,motifdata,excludepromoters, onlymax,weight)), col = list( condition = c("lps" = "#0066CC", "dexlps" = "#339966", "DexLPS-LPS" = "orange"), motifdata=c("summitregion" = "purple", "abcregion" = "chocolate4"), excludepromoters= c("FALSE"=myreds[1], "onlyNONself"=myreds[2], "all"=myreds[3]), onlymax=c("prox"="black", "TRUE"="lightgrey", "FALSE"="darkgrey"), weight=c("FALSE"="lightgoldenrod", "abcscore"="goldenrod") ), AUC = anno_barplot(AUC_metrics %>% filter(names %in% names(model_coefs_numeric)) %>% arrange(match(names,names(model_coefs_numeric))) %>% dplyr::pull(net_test), ylim=c(0.5,0.8)), show_annotation_name = c(FALSE,FALSE, FALSE,FALSE,FALSE,TRUE), show_legend = c(group=TRUE), annotation_legend_param = list( labels_gp = gpar(fontsize=6), title_gp = gpar(fontsize=6) ), simple_anno_size = unit(2,"mm") ) mycolorramp <- circlize::colorRamp2(breaks=c(-1.5,0,1.5), colors=c("blue","white","red")) heatmap <- Heatmap( col=mycolorramp, model_coefs_numeric, top_annotation = col_ha, right_annotation = row_ha, row_split = 4, column_title="Model coefficients", column_title_gp = gpar(fontsize = 6, fontface = "bold"), column_names_gp = gpar(fontsize = 6), row_names_gp = gpar(fontsize = 6), name="z-scaled coefficient", #Title on top of legend clustering_distance_rows = "euclidean", clustering_method_rows = "ward.D2", clustering_distance_columns = "euclidean", clustering_method_columns = "ward.D2", show_column_dend = TRUE, show_row_dend = TRUE, cluster_rows = TRUE, cluster_columns = TRUE, show_row_names = TRUE, show_column_names = FALSE, row_names_side = "left", column_dend_height = unit(0.5, "cm"), heatmap_legend_param = list( title_position = "leftcenter-rot", legend_direction="vertical", labels_gp = gpar(fontsize=6), title_gp = gpar(fontsize=6), legend_height = unit(1, "cm"), at = c(-1.5, 0, 1.5), labels = c("-1.5", "0", "1.5") ) ) return ( grid.grabExpr( draw(heatmap, heatmap_legend_side="right", merge_legend = TRUE) ) ) } gg_model_coefs_joint_heatmap <- make_coefficient_heatmap(model_coefs_joint) #--------------------------------------------- #---- coefficients summitranges #--------------------------------------------- make_summitranges_coefficients_heatmap <- function(model_coefs){ #assign any 0 coefficients an NA (so they don't get affected by scaling) model_coefs[model_coefs==0] <- NA model_coefs <- model_coefs %>% #filter(featurename!="(Intercept)") %>% # remove the intercept term for plotting dplyr::select("featurename","prox" , contains("summitregion")) # select only models based on summitregion # count how many models have the motif included with a non-zero coef model_coefs_ftr <- model_coefs %>% mutate( sum = rowSums(across(where(is.numeric)), na.rm = TRUE)) %>% mutate( motifhascoef_counts = rowSums(!is.na(.)) ) %>% mutate(., across(3:(ncol(.)-2), ~(scale(.) %>% as.vector))) %>% #scale before filtering on spec. factors filter( motifhascoef_counts >= 15 ) #filter(prox!=0) # remove extra columns before passing it to heatmap model_coefs_numeric <- model_coefs_ftr %>% tibble::column_to_rownames("featurename") %>% dplyr::select(!c( sum, motifhascoef_counts)) row_ha = rowAnnotation( counts = anno_barplot(model_coefs_ftr$motifhascoef_counts) ) # take annotation from AUC_metrics #create annotation labels for the heatmap myreds <- brewer.pal(3,"OrRd") col_ha <- HeatmapAnnotation(df = AUC_metrics %>% filter(names %in% names(model_coefs_numeric)) %>% arrange(match(names,names(model_coefs_numeric))) %>% dplyr::select(c(condition,motifdata,excludepromoters, onlymax,weight)), col = list( condition = c("lps" = "#0066CC", "dexlps" = "#339966", "DexLPS-LPS" = "orange"), motifdata=c("summitregion" = "purple", "abcregion" = "chocolate4"), excludepromoters= c("FALSE"=myreds[1], "onlyNONself"=myreds[2], "all"=myreds[3]), onlymax=c("prox"="black", "TRUE"="lightgrey", "FALSE"="darkgrey"), weight=c("FALSE"="lightgoldenrod", "abcscore"="goldenrod") ), AUC = anno_barplot(AUC_metrics %>% filter(names %in% names(model_coefs_numeric)) %>% arrange(match(names,names(model_coefs_numeric))) %>% dplyr::pull(net_test), ylim=c(0.5,0.8)), show_annotation_name = c(FALSE,FALSE, FALSE,FALSE,FALSE,TRUE), show_legend = FALSE, annotation_legend_param = list( labels_gp = gpar(fontsize=6), title_gp = gpar(fontsize=6) ), simple_anno_size = unit(2,"mm") ) mycolorramp <- circlize::colorRamp2(breaks=c(-1.5,0,1.5), colors=c("blue","white","red")) heatmap <- Heatmap( col=mycolorramp, model_coefs_numeric, top_annotation = col_ha, right_annotation = row_ha, row_split = 4, column_title="Model coefficients", column_title_gp = gpar(fontsize = 6, fontface = "bold"), column_names_gp = gpar(fontsize = 6), row_names_gp = gpar(fontsize = 6), name="z-scaled coefficient", #Title on top of legend clustering_distance_rows = "euclidean", clustering_method_rows = "ward.D2", clustering_distance_columns = "euclidean", clustering_method_columns = "ward.D2", show_column_dend = TRUE, show_row_dend = TRUE, cluster_rows = TRUE, cluster_columns = TRUE, show_row_names = TRUE, show_column_names = FALSE, row_names_side = "left", column_dend_height = unit(0.5, "cm"), show_heatmap_legend = FALSE ) return ( grid.grabExpr( draw(heatmap, heatmap_legend_side="right", merge_legend = TRUE) ) ) } gg_model_summitregioncoefs_joint_heatmap <- make_summitranges_coefficients_heatmap(model_coefs_joint) #--------------------------------------------- #---- compare model coefficients of best models #--------------------------------------------- # check in AUC what is the best model max_net_test$names #"motifcounts_abcregion_condition_DexLPS-LPS_exclprom_onlyNONself_onlymax_FALSE_sepPromEnh_FALSE_weight_FALSE" # keep those rows where at least one of the models has a non-zero coefficient keepRows <- which(rowSums(model_coefs_joint [,c("prox",max_net_test$names)]) != 0 ) model_coefs_long <- model_coefs_joint[keepRows,] %>% dplyr::select(c(featurename,prox,max_net_test$names)) %>% dplyr::rename(abc=max_net_test$names)%>% tidyr::pivot_longer(., col=c('prox','abc'), names_to = "model", values_to = "coefficients") model_coefs_long <- model_coefs_long %>% mutate(my_color = case_when(coefficients>0 & model=="prox" ~ "darkred", coefficients<0 & model=="prox" ~ "darkblue", coefficients>0 & model!="prox" ~ "lightred", coefficients<0 & model!="prox" ~ "lightblue" )) gg_bestmodels <- ggplot(data=model_coefs_long)+ geom_bar(aes(x=reorder(featurename,-coefficients), y=coefficients, fill=model),stat="identity", position = position_dodge(width = 0.5),width=0.5)+ scale_fill_manual(values=c("prox"="purple", "abc" = "brown"), labels=c("prox. based", "ABC based"))+ labs(x="motifname", y="coefficient")+ theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1), legend.position = c(0.8, 0.8)) gg_bestmodels #--------------------------------------------- #---- or just display the prox based one #--------------------------------------------- keepRows <- which(model_coefs_joint [,"prox"] != 0 ) model_coefs_data <- model_coefs_joint[keepRows,] %>% mutate(my_color = case_when(prox>0 ~ "positive", prox<0 ~ "negative")) gg_proxmodel <- ggplot(data= model_coefs_data)+ geom_bar(aes(x=reorder(featurename,-prox), y=prox, fill=my_color), stat="identity", position = position_dodge(width = 0.5),width=0.5)+ scale_fill_manual(values=c("positive"="red", "negative" = "blue"))+ labs(x="motifname", y="coefficient", fill="coefficient sign")+ theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust=1), legend.position = c(0.8,0.8), panel.grid.major = element_blank()) gg_proxmodel #--------------------------------------------- #---- with red and blue, paired dark and light for the 2 models #--------------------------------------------- #my_pal <- RColorBrewer::brewer.pal(6,"Paired") #gg_bestmodels <- ggplot(data=model_coefs_long)+ # geom_bar(aes(x=featurename, y=coefficients, fill=my_color, colour=model),stat="identity", position = position_dodge(width = 0.5),width=0.5)+ # scale_fill_manual(values=c("darkred"=my_pal[6], # "darkblue"=my_pal[2], # "lightred"=my_pal[5], # "lightblue"=my_pal[1]), # labels=c("only peakregions\n- proxbased - up", # "only peakregions\n- proxbased - down", # "all enhancers\n- abcbased - up", # "all enhancers\n- abcbased - down" # ))+ # scale_colour_manual(values=c("prox"="red", # "motifcounts_peak_condition_dexlps_exclprom_all_onlymax_FALSE_sepPromEnh_FALSE_weight_abcscore" = "blue"), # labels=c("prox. based", # "ABC based"))+ # theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))+ # guides(fill="none") #--------------------------------------------- #---- load raw counts ggplot #--------------------------------------------- gg_rawcounts <- readRDS(opt$raw_counts) #--------------------------------------------- #---- put figure together #--------------------------------------------- gg_coef_heatmaps <- ggpubr::ggarrange(gg_model_coefs_joint_heatmap, gg_model_summitregioncoefs_joint_heatmap, labels = c("B", "C"), ncol = 2, nrow = 1, widths=c(1.2,1) ) gg_coef_heatmaps #gg_bottomrow <- ggpubr::ggarrange(gg_proxmodel, gg_rawcounts, # labels = c("D", "E"), # ncol = 2, nrow = 1, widths=c(2,1) #) #gg_bottomrow #full_panel <- ggpubr::ggarrange(gg_AUC, gg_coef_heatmaps, gg_bottomrow, full_panel <- ggpubr::ggarrange(gg_AUC, gg_coef_heatmaps, gg_bestmodels, labels = c("A", NA, "D"), nrow=3, heights=c(0.5,1,0.3)) full_panel ggsave(here("./results/current/Figures/Figure_GLMs.png"), full_panel, width=190, height=240, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_GLMs.pdf"), full_panel, width=190, height=240, units="mm", bg="white") #--------------------------------------------- #---- save models with enhancers and promoters as separate features #--------------------------------------------- # possible supplemental: gg_model_coefs_sep_heatmap <- make_coefficient_heatmap(model_coefs_sep) gg_model_coefs_sep_heatmap gg <- ggpubr::ggarrange(gg_model_coefs_sep_heatmap, labels = c(NA), ncol = 1, nrow = 1, widths=c(1)) gg ggsave(here("./results/current/Figures/Figure_GLMs_coefssep.png"), gg, width=190, height=190, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_GLMs_coefssep.pdf"), gg, width=190, height=190, units="mm", bg="white") |
R
ggplot2
dplyr
optparse
RColorBrewer
ggpubr
here
ComplexHeatmap
circlize
From
line
1
of
scripts/figure_GLMs.r
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--summitAnno_expr"), type="character", help="Path to file with GR idr summits annotated to expressed genes (as anno object)"), make_option(c("--summitAnno_df_expr"), type="character", help="Path to file with GR idr summits annotated to expressed genes (as dataframe)"), make_option(c("--permtest_res"), type="character", help="Path to rds file with results of permutation test of group differences"), make_option(c("--fimo_results"), type="character", help="Path to rds file with fimo hits within summitregions (1000bp)"), make_option(c("--chipseq_summit_granges"), type="character", help="Path to summit file of IDR peaks"), make_option(c("--deeptools"), type="character", help="Path to png of deeptoolsheatmap split by activating vs repressing peaks"), make_option(c("--streme_100bp_up"), type="character", help="Path to streme xml file for 100bp regions around activating peak regions"), make_option(c( "--streme_100bp_down"), type="character", help="Path to streme xml file for 100bp regions around repressing peak regions")) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(memes, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(universalmotif, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(png, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=8, family = "ArialMT", colour="black"), title=element_text(size=10, family="ArialMT", colour="black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text = element_text(size=8, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(8, 'points'), #change legend key size legend.key.height = unit(8, 'points'), #change legend key height legend.key.width = unit(8, 'points'), #change legend key width legend.text = element_text(size=8, family="ArialMT", colour="black") ) #------------------------------- ## 0. load data #------------------------------- summitAnno_expr <- readRDS( opt$summitAnno_expr) summitAnno_df_expr <- readRDS( opt$summitAnno_df_expr ) permtest_res <- readRDS( opt$permtest_res ) fimo_results <- readRDS( opt$fimo_results ) ChIPseq_summit_Granges <- readRDS( opt$chipseq_summit_granges ) # Take 100bp windows around ChIP-seq summits summit_flank_100bp <- ChIPseq_summit_Granges %>% plyranges::anchor_center() %>% plyranges::mutate(width = 100) # Take 100bp windows around ChIP-seq summits summit_flank_1000bp <- ChIPseq_summit_Granges %>% plyranges::anchor_center() %>% plyranges::mutate(width = 1000) #--------------------------------- #### Peak distance from TSS #--------------------------------- #Turning things around, let's check #* how far the closest peak is from the TSS of up-vs downregulated gene #* how far we need to go from the TSS to have 2 or 3 peaks mapping to the gene distbygene <- summitAnno_df_expr %>% group_by(SYMBOL, change) %>% summarise(min_dist=min(abs(distanceToTSS)), mean_dist=mean(abs(distanceToTSS))) %>% ungroup() %>% mutate(logmindist=log2(min_dist+1)) ggplot( ) + geom_histogram(data=distbygene %>% filter(change=="up"), aes(x=min_dist, fill="up"), alpha=.3, binwidth = 5000) + geom_histogram(data=distbygene %>% filter(change=="down"), aes(x=min_dist, fill="down"), alpha=.3, binwidth = 5000) + expand_limits(x=c(0,400000))+ labs(title="Dist. to nearest peak: split by direction of expression change")+ scale_x_continuous(breaks = seq(0, 400000, by = 100000), labels = paste(seq(0, 400000, by = 100000) / 1000,"kb"))+ scale_fill_manual(name = "", values = c( "up" = "red", "down" = "blue")) # Zoom into reasonable region gg_distbychange <- ggplot( ) + geom_histogram(data=distbygene %>% filter(change=="ns"), aes(x=logmindist,y=..density.., fill="ns"), binwidth=0.5, alpha=.2) + geom_histogram(data=distbygene %>% filter(change=="up"), aes(x=logmindist,y=..density.., fill="up"), binwidth=0.5, alpha=.2) + geom_histogram(data=distbygene %>% filter(change=="down"), aes(x=logmindist,y=..density.., fill="down"), binwidth=0.5, alpha=.2) + geom_density(data=distbygene %>% filter(change=="up"), aes(x=logmindist, colour="up"), alpha=.3, show.legend = FALSE) + geom_density(data=distbygene %>% filter(change=="down"), aes(x=logmindist, colour="down"), alpha=.3, show.legend = FALSE) + geom_density(data=distbygene %>% filter(change=="ns"), aes(x=logmindist, colour="ns"), alpha=.3, show.legend = FALSE) + geom_vline(xintercept = log2(30000), size=1, colour="black", linetype=2)+ scale_x_continuous(breaks = c( log2(0+1), log2(1000+1), log2(5000+1), log2(10000+1), log2(30000+1),log2(100000+1) ), labels = paste(c(0,1000,5000,10000,30000,100000) / 1000,"kb"))+ scale_fill_manual(name = "", values = c( "up" = "red", "down" = "blue", "ns" = "black"))+ scale_colour_manual(name = "", values = c( "up" = "red", "down" = "blue", "ns" = "black"))+ guides(colour="none")+ theme(axis.text.x = element_text(angle=45, vjust=1, hjust=1), legend.position = c(0.15,0.8))+ labs(x="Distance to nearest peak - genecentric") gg_distbychange #--------------------------------- # --- load deeptools figure #--------------------------------- deeptools <- png::readPNG( opt$deeptools ) gg_deeptools <- ggplot() + ggpubr::background_image(deeptools) + # so it doesn't get squished #coord_fixed()+ # This ensures that the image leaves some space at the edges theme(plot.margin = margin(t=0, l=0, r=0, b=0, unit = "cm"), axis.line = element_blank()) #-------------------------------------- #- permutations #-------------------------------------- groupdiff <- as.numeric(permtest_res[[3]]) gg_perm_dist <- ggplot()+ geom_histogram(aes(x=permtest_res[[2]]), bins=100)+ labs(title="", x="Permutation based \n group differences")+ xlim(-10000,10000)+ geom_vline( xintercept = groupdiff, size=1, colour="red", linetype=1) #-------------------------------------- #- memes #-------------------------------------- streme_up_results <- memes::importStremeXML( opt$streme_100bp_up ) streme_up_results <- streme_up_results %>% mutate(name = paste(consensus, pval)) gg_streme_up <- streme_up_results[1:5,] %>% to_list() %>% view_motifs(text.size = 7, tryRC = TRUE) #theme(plot.margin = margin(0,0,2.4,0, "cm")) gg_streme_up streme_down_results <- memes::importStremeXML( opt$streme_100bp_down ) streme_down_results <- streme_down_results %>% mutate(name = paste(consensus, pval)) gg_streme_down <- streme_down_results[1:5,] %>% to_list() %>% view_motifs(text.size = 7, tryRC = TRUE) gg_streme_down #--------------------------------- # --- fimo chi-square #--------------------------------- make_fimo_chisquare_plot <- function(summit_flank, fimo_results, seqwidth){ input_intersect_hits <- summit_flank %>% plyranges::join_overlap_intersect(fimo_results) fimo_counts <- as.data.frame(input_intersect_hits) %>% group_by(motif_alt_id,directionchange) %>% summarize(count=n()) fimo_counts <- fimo_counts %>% tidyr::pivot_wider( names_from = directionchange, values_from = count) tbl_directionchange <- table(summit_flank$directionchange) setsize <- tbl_directionchange[["down"]] + tbl_directionchange[["up"]] fimo_counts <- fimo_counts %>% na.omit() fimo_counts <- fimo_counts %>% rowwise() %>% mutate( test_stat = chisq.test(c(down, up), p=c(tbl_directionchange[["down"]] / setsize, tbl_directionchange[["up"]] / setsize) )$statistic, p_val = chisq.test(c(down, up), p=c(tbl_directionchange[["down"]] / setsize, tbl_directionchange[["up"]] / setsize) )$p.value ) fimo_counts$p_adj <-p.adjust(fimo_counts$p_val, method="fdr") # available methods: c("holm", "hochberg", "hommel", "bonferroni", "BH", "BY", "fdr", "none") # plot fimo results #------------------------------------------------ colnames(fimo_counts) <- c("motif_name","counts_down","ns","counts_up","test_stat","p_val","p_adj") fimo_counts <- fimo_counts %>% mutate( categories = case_when(grepl("NR3C", motif_name, ignore.case = TRUE) | motif_name=="Ar" | motif_name=="Nr2F6" ~ "NR", grepl("NFKB", motif_name, ignore.case = TRUE) | motif_name=="REL" | motif_name=="RELA" | motif_name=="RELB" ~ "NFKB", grepl("CREB", motif_name, ignore.case = TRUE) | motif_name=="Atf1"| motif_name=="CREM" ~ "CREB", grepl("POU", motif_name, ignore.case = TRUE) ~ "POU", grepl("KLF", motif_name, ignore.case = TRUE) ~ "KLF", grepl("JDP", motif_name, ignore.case = TRUE) ~ "AP-1", grepl("JUN", motif_name, ignore.case = TRUE) ~ "AP-1", grepl("IRF", motif_name, ignore.case = TRUE) ~ "IRF", grepl("Stat", motif_name, ignore.case = TRUE) ~ "STAT", TRUE ~ "Other") ) # compute fit and plot regression line fit <- lm(fimo_counts$counts_up ~ fimo_counts$counts_down, data = fimo_counts) gg <- ggplot(fit$model, aes_string(x = names(fit$model)[2], y = names(fit$model)[1])) + geom_point(colour="grey") + stat_smooth(method = "lm", col = "blue") + ggrepel::geom_label_repel(data= fimo_counts %>% dplyr::arrange(p_adj) %>% head(20) %>% dplyr::filter(counts_down<counts_up) , aes(x=counts_down, y=counts_up, label=motif_name, colour=categories), size=2, min.segment.length = 0, position = ggrepel::position_nudge_repel(x=-(seqwidth/2), y=(seqwidth/2)), label.padding = 0.1, box.padding = 0.1, show.legend=FALSE)+ ggrepel::geom_label_repel(data= fimo_counts %>% dplyr::arrange(p_adj) %>% head(20) %>% dplyr::filter(counts_down>counts_up) , aes(x=counts_down, y=counts_up, label=motif_name, colour=categories), size=2, min.segment.length = 0, force_pill=0.5, force = 5, max.overlaps = 50, position = ggrepel::position_nudge_repel(x=(seqwidth/2), y=-(seqwidth/2)), label.padding = 0.1, box.padding = 0.1, show.legend=FALSE)+ #highlight the same motifs by coloring the point geom_point(data= fimo_counts %>% dplyr::arrange(p_adj) %>% head(20), aes(x=counts_down, y=counts_up, colour=categories))+ scale_colour_manual(name="Motif family", values=c("NR"="firebrick","NFKB"="darkblue","IRF"="seagreen","KLF"="deeppink", "STAT"="darkolivegreen3","POU"="coral","Other"="black"))+ expand_limits(x=-(seqwidth/2), y=-(seqwidth/2))+ # use the slop to dynamically code this (when windows around summit are smaller, we don't have overplotting issues in the lower left) labs(title=paste0("Peakregion ",seqwidth," bp."), x="#Motifmatches in peaks of downregulated genes", y="#Motifmatches in peaks of upregulated genes", subtitle = paste("Adj R2 = ",signif(summary(fit)$adj.r.squared, 2) ) ) + theme(legend.position = "bottom") results=list() results[["fimo_counts"]] <- fimo_counts results[["plot"]] <- gg return(results) } fimo_100_results <- make_fimo_chisquare_plot(summit_flank_100bp,fimo_results, 100) gg_fimo_100 <- fimo_100_results[["plot"]] fimo_100_results[["fimo_counts"]] %>% arrange(p_adj) %>% head(n=20) fimo_1000_results <- make_fimo_chisquare_plot(summit_flank_1000bp,fimo_results, 1000) gg_fimo_1000 <- fimo_1000_results[["plot"]] #undebug(make_fimo_chisquare_plot) #--------------------------------- # --- merge figures into panel #--------------------------------- gg_r1 <- ggpubr::ggarrange(gg_distbychange, gg_perm_dist , labels = c("A","B"), widths = c(1,0.6), ncol = 2, nrow=1) gg_r2 <- ggpubr::ggarrange(gg_streme_up, gg_streme_down, labels = c("D","E"), widths = c(1,1), ncol = 2, nrow=1) gg_c1 <- ggpubr::ggarrange(gg_r1, gg_r2, labels = c(NA, NA), ncol = 1, nrow=2, heights=c(1,1)) full_top <- ggpubr::ggarrange(gg_c1, gg_deeptools, labels = c(NA, "C"), ncol = 2, nrow=1, widths = c(1.7,1)) gg_bottomrow <- ggpubr::ggarrange(gg_fimo_100, gg_fimo_1000 , labels = c("F","G"), widths = c(1,1), ncol = 2, nrow=1) full_panel <- ggpubr::ggarrange(full_top, gg_bottomrow, labels = c(NA, NA), nrow = 2, ncol=1, heights = c(1.7,1)) full_panel ggsave(here("./results/current/Figures/Figure_peakgeneannotation.png"), full_panel, width=190, height=200, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_peakgeneannotation.pdf"), full_panel, width=190, height=200, units="mm", bg="white") |
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--log2fcthresh"), type="numeric", help="Log2FC threshold used in addition to adj.pval to define significant genes"), make_option(c("--chipseq_summits"), type="character", help="Path to summit file of IDR peaks"), make_option(c("--genekey"), type="character", help="Path to biomart genekey that mapps ensembl geeneIDs to MGI symbols"), make_option(c("--contrast_DexVSDexLPS"), type="character", help="Path to annotated tsv file of DeSeq2 contrast of DexLPS vs LPS"), make_option(c("--meme_db_path"), type="character", help="Path to JASPAR motif db file"), make_option(c( "--rna_nascent_fpkm"), type="character", help="FPKM matrix of 4sU experiment"), make_option(c("-o", "--outdir"), type="character", help="Path to output directory")) opt <- parse_args(OptionParser(option_list=option_list)) # set output for logfile to retrieve stats for plot later sink(file=paste0(opt$outdir,"figure_proxanno_prepdata.out")) suppressPackageStartupMessages(library(memes, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(universalmotif, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(biomaRt, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(TxDb.Mmusculus.UCSC.mm10.knownGene, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(BSgenome.Mmusculus.UCSC.mm10, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ChIPseeker, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(stringr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) #------------------------------- ## Import references #------------------------------- # for gene annotation txdb <- TxDb.Mmusculus.UCSC.mm10.knownGene # for the sequence # either use masked or unmasked (the mask does NOT seem to be for repeats though) mm.genome <- BSgenome.Mmusculus.UCSC.mm10 #------------------------------- ### Determine expressed genes #------------------------------- rna_nascent <- read.table(opt$rna_nascent_fpkm, header=TRUE) print("Determine expressed genes using 4sU data") # what does the read count distribution look like? # compute median per gene and plot it as histogram rna_nascent$median_genecounts <- apply(rna_nascent[,-1], 1, FUN=median) # lots of medians that are below 1 hist(log10(rna_nascent$median_genecounts)) # filter based on expression expressed_genes <- rna_nascent %>% dplyr::filter(median_genecounts > 0) #------------------------------- ### Load genekey to annotate ensembl to mgi #------------------------------- geneKey <- read.delim(opt$genekey) #merge gene annotations to results table expressed_genes <- merge(expressed_genes, geneKey, by.x="Geneid", by.y="ensembl_gene_id") #------------------------------- ## Getting the sequences #------------------------------- # import summit of ChIPseq peaks ChIPseq_summits <- read.table(opt$chipseq_summits) ChIPseq_ranges <- GRanges(seqnames = ChIPseq_summits[,c("V1")], ranges = IRanges(start=ChIPseq_summits[,c("V2")], end=ChIPseq_summits[,c("V3")]-1)) # to make up for 0 vs 1 encoding ChIPseq_ranges$id <- c(1:length(ChIPseq_ranges)) # NOTE: if we only want to annotate to genes that are expressed, we could use ChIPpeakAnno and a filtered annoDB object instead # annotate it to genes print("Annotate ChIPseq summit to closest gene (using genomic reference)") summitAnno <- annotatePeak(ChIPseq_ranges, tssRegion=c(-3000, 3000), TxDb=txdb, annoDb = "org.Mm.eg.db") summitAnno_df <- summitAnno %>% as.data.frame() # see which ones are DE genes and add that info to GRanges as column "directionchange" print("Add info on which genes are DE to the annotated summits") DE_4sU <- read.delim(opt$contrast_DexVSDexLPS) summitAnno_df <- left_join(summitAnno_df, DE_4sU[,c("mgi_symbol","padj","log2FoldChange")], by = c("SYMBOL" = "mgi_symbol")) summitAnno_df <- summitAnno_df %>% mutate(change = case_when(padj<0.05 & log2FoldChange > opt$log2fcthresh ~ "up", padj<0.05& log2FoldChange < -opt$log2fcthresh ~"down", TRUE ~ "ns") ) # save info on gene annotation ChIPseq_ranges$mgi_symbol[match(summitAnno_df$id, ChIPseq_ranges$id )] <- summitAnno_df$SYMBOL # assign directionchange as metadata column ChIPseq_ranges$directionchange[match(summitAnno_df$id, ChIPseq_ranges$id )] <- summitAnno_df$change # ass distance to TSS ChIPseq_ranges$distanceToTSS[match(summitAnno_df$id, ChIPseq_ranges$id )] <- summitAnno_df$distanceToTSS #------------------------------- ## Prefilter motifdb to motifs that are expressed in celltype #------------------------------- print("Prefilter meme_db for those motifs expressed in our 4sU data") meme_db <- read_meme(opt$meme_db_path) %>% to_df() meme_db_expressed <- meme_db %>% # the altname slot of meme_db contains the gene symbol (this is database-specific) # avoid mismatches cased by casing and keep motif if at least one part of composite is expressed tidyr::separate(altname, into=c("tf1", "tf2"), sep="::",remove=FALSE) %>% filter( str_to_upper(tf1) %in% str_to_upper(expressed_genes$mgi_symbol) | str_to_upper(tf2) %in% str_to_upper(expressed_genes$mgi_symbol)) %>% # we don't need the split TF info downstream dplyr::select(!c("tf1","tf2")) print("Number of motifs pre-filtering: ") nrow(meme_db) print("Number of motifs post-filter: ") nrow(meme_db_expressed) #------------------------------- ## OPTIONAL: only run with motifs of interest #------------------------------- meme_motifsOI <- meme_db_expressed %>% filter( grepl("STAT", str_to_upper(altname)) | grepl("NR3C", str_to_upper(altname)) ) #------------------------------- ## FIGURES on peak gene annotation #------------------------------- # filter peaks for those annotated to genes that are expressed summitAnno_expr <- subset(summitAnno, summitAnno@anno$SYMBOL %in% expressed_genes$mgi_symbol) # filter the df version in the same fashion summitAnno_df_expr <- summitAnno_df %>% filter(SYMBOL %in% expressed_genes$mgi_symbol) #--------------------------------- # --- some stats #--------------------------------- distbygene <- summitAnno_df_expr %>% group_by(SYMBOL, change) %>% summarise(min_dist=min(abs(distanceToTSS)), mean_dist=mean(abs(distanceToTSS))) %>% ungroup() %>% mutate(logmindist=log2(min_dist+1)) # why 30kb cutoff distbygene_allDE <- distbygene %>% filter(!change=="ns") %>% mutate(change=factor(change,levels=c("down","up"))) print("We need to justify why we picked a cutoff of 30kb.") print("From a genecentric perspective, we want to include the peak regions that most likely have a regulating function on the gene.") print("With a cutoff of 30kb, how many genes DONT have at least one peak within that range?") tbl <- table(distbygene_allDE$min_dist > 30000) tbl[2]/(tbl[1]+tbl[2]) print("How many genes do we lose of both sets by using that cutoff?") print("In the upregulated fraction:") table( (distbygene %>% filter(change=="up"))$min_dist > 30000) print("In the downregulated fraction:") table( (distbygene %>% filter(change=="down"))$min_dist > 30000) print("Min and mean dist for the genes with log2FC >", opt$log2fcthresh) distbygene %>% filter(change=="up") %>% summarise_all(mean) %>% print() print("Min and mean dist for the genes with log2FC <", opt$log2fcthresh ) distbygene %>% filter(change=="down") %>% summarise_all(mean) %>% print() print("How many peaks per UPregulated gene:") summitAnno_df_expr %>% filter(change=="up")%>% filter(abs(distanceToTSS)<30000) %>% group_by(SYMBOL) %>% summarise(count=n()) %>% pull(count) %>% mean() print("How many peaks per DOWNregulated gene:") summitAnno_df_expr %>% filter(change=="down")%>% filter(abs(distanceToTSS)<30000) %>% group_by(SYMBOL) %>% summarise(count=n()) %>% pull(count) %>% mean() print("The distances between the peaks mapping to the same gene.") print("returns NA if only one 1 peak is annotated to the gene - those are excluded") print("For upregulated genes:") summitAnno_df_expr %>% filter(change=="up")%>% filter(abs(distanceToTSS)<30000) %>% group_by(SYMBOL) %>% summarise(meanpeakdist = mean(dist(distanceToTSS))) %>% # filter(!is.na(meanpeakdist)) %>% pull(meanpeakdist) %>% mean() print("For downregulated genes:") summitAnno_df_expr %>% filter(change=="down")%>% filter(abs(distanceToTSS)<30000) %>% group_by(SYMBOL) %>% summarise(meanpeakdist = mean(dist(distanceToTSS))) %>% filter(!is.na(meanpeakdist)) %>% pull(meanpeakdist) %>% mean() #-------------------------------------- #- permutations #-------------------------------------- # Difference in means groupdiff <- diff(tapply(distbygene_allDE$min_dist, distbygene_allDE$change, mean)) print("The mean minimum distance is smaller for the upregulated set") print(paste("The group difference is: ",groupdiff)) print("Permutation test to see if this difference between the groups is meaningful") #Permutation test permutation.test <- function(group, outcome, n, reference){ distribution=c() result=0 for(i in 1:n){ distribution[i]=diff(by(outcome, sample(group, length(group), FALSE), mean)) } result=sum(abs(distribution) >= abs(groupdiff))/(n) return(list(result, distribution, groupdiff)) } permtest_res <- permutation.test(distbygene_allDE$change, distbygene_allDE$min_dist, 100000, groupdiff) #-------------------------------------- #- export objects #-------------------------------------- #--------------------------------- # --- export results of the permutation test saveRDS(permtest_res, file=paste0(opt$outdir,"permtest_res.rds")) #--------------------------------- # --- export up and downregulated summitfraction for deeptools table(ChIPseq_ranges$directionchange) dir.create(paste0(opt$outdir,"peaks_annot2DEgenes_30kb_log2FC0.58/")) export.bed(ChIPseq_ranges %>% filter(directionchange == "up") %>% filter(abs(distanceToTSS)<30000) , con=paste0(opt$outdir,"peaks_annot2DEgenes_30kb_log2FC0.58/UP_summit_unmerged.bed")) export.bed(ChIPseq_ranges %>% filter(directionchange == "down") %>% filter(abs(distanceToTSS)<30000) , con=paste0(opt$outdir,"peaks_annot2DEgenes_30kb_log2FC0.58/DOWN_summit_unmerged.bed")) #------------------------------- ## export objects to run memes afterwards saveRDS(ChIPseq_ranges, file=paste0(opt$outdir,"../memes_bioc/ChIPseq_summit_Granges.rds")) saveRDS(meme_db_expressed, file=paste0(opt$outdir,"../memes_bioc/meme_db_4sUexpressed.rds")) #------------------------------- ## export objects for ggplot figures saveRDS(summitAnno, file=paste0(opt$outdir,"summitAnno.rds")) saveRDS(summitAnno_expr, file=paste0(opt$outdir,"summitAnno_expr.rds")) saveRDS(summitAnno_df_expr, file=paste0(opt$outdir,"summitAnno_df_expr.rds")) sink() |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--tfactivity"), type="character", help="TF activity, computed using macrophage specific TSS"), make_option(c("--expr"), type="character", help="Path to file with normalized counts, aggregated per mgi symbol"), make_option(c("--difffootprint"), type="character", help="Path to txt file with differential statistics from footprinting analysis"), make_option(c("--memedb_expressed"), type="character", help="Path to rds file of memedb motifs filtered for those expressed"), make_option(c("--heatmap"), type="character", help="Path to deeptools heatmap of motif of interest"), make_option(c("--chipms"), type="character", help="Path to xlsx file with statistics on GR-ChIPMS analysis") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(universalmotif, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "white", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=10, family = "ArialMT", colour="black"), title=element_text(size=8, family="ArialMT", colour="black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text = element_text(size=8, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(8, 'points'), #change legend key size legend.key.height = unit(8, 'points'), #change legend key height legend.key.width = unit(8, 'points'), #change legend key width legend.text = element_text(size=8, family="ArialMT", colour="black") ) #------------------------------------------------------------------------------- # Load input files #------------------------------------------------------------------------------- meme_db_expressed <- readRDS(opt$memedb_expressed) expr <- read.delim(opt$expr) TFA <- read.delim(opt$tfactivity) footprint <- read.table(opt$difffootprint, header=TRUE) #------------------------------------------------------------------------------- # Differential footprints #------------------------------------------------------------------------------- footprint <- footprint %>% mutate(motif_name = stringr::str_split_fixed(Motif, "\\.",3)[,3], Protection_score_diff = Protection_Score_DexLPS - Protection_Score_LPS, categories = case_when(grepl("NR3C", motif_name, ignore.case = TRUE) | motif_name=="Ar" | motif_name=="Nr2F6" ~ "NR", grepl("CREB", motif_name, ignore.case = TRUE) | motif_name=="Atf1"| motif_name=="CREM" ~ "CREB", grepl("KLF", motif_name, ignore.case = TRUE) ~ "KLF", grepl("ATF", motif_name, ignore.case = TRUE) ~ "AP-1", grepl("JDP", motif_name, ignore.case = TRUE) ~ "AP-1", grepl("JUN", motif_name, ignore.case = TRUE) ~ "AP-1", grepl("IRF", motif_name, ignore.case = TRUE) ~ "IRF", TRUE ~ "Other"), padj = p.adjust(P_values, method = "fdr") ) # filter on those that we tested in the GLMs # even if we do, FDR correction kills almost all significance #footprint_ftr <- footprint %>% # filter(motifnames %in% meme_db_expressed$altname) %>% # mutate( padj = p.adjust(P_values, method = "fdr")) #should we additionally filter on the protection score or not? If we argue for transient binding, we might not want to table(footprint %>% filter(Num>100 & P_values < 0.05) %>% dplyr::pull(categories)) gg_difffootprint <- ggplot()+ geom_point(data = footprint %>% filter(Num>100 ), aes(x=Num, y=-log10(P_values)), colour="grey") + geom_point(data = footprint %>% filter(Num>100 & P_values < 0.05), aes(x=Num, y=-log10(P_values), colour=categories))+ ggrepel::geom_label_repel(data = footprint %>% filter(Num>100 & P_values < 0.05 & categories=="Other") , aes(x=Num, y=-log10(P_values), label=motif_name, colour=categories), size=2.5, min.segment.length = 0, position = ggrepel::position_nudge_repel(x=8000, y=0.1), label.padding = 0.1, box.padding = 0.1, show.legend = FALSE)+ scale_colour_manual(name="Motif family", values=c("NR"="firebrick", "AP-1"="dodgerblue", "CREB"="purple", "IRF"="seagreen", "KLF"="deeppink", "Other"="black"))+ ylim(c(0,3)) gg_difffootprint #footprint %>% filter(Num>100 & Protection_Score_DexLPS>1) %>% arrange(P_values) %>% View() #footprint %>% arrange(desc(abs(Protection_score_diff))) %>% head(20) # requires poppler-cpp system installation #magick::image_read_pdf(here("../mac_atacseq/results/current/footprints/DexLPSvLPS_diff_footprint/Lineplots/MA1127.1.FOSB::JUN.pdf")) gg_placeholder <- ggplot() + theme(plot.margin = margin(t=0, l=0, r=0, b=0, unit = "cm"), axis.line = element_blank()) #------------------------------------------------------------------------------- # Expression levels of STATs #------------------------------------------------------------------------------- expr_long <- expr %>% tibble::rownames_to_column(var="mgi_symbol") %>% tidyr::pivot_longer(cols=-c(mgi_symbol), # everything except mgi_symbol column names_to="sample") %>% tidyr::extract( "sample", c("condition","rep","type"), regex = "(.*)([0-9])_([^_]+)$") %>% mutate(condition=factor(condition, levels=c("V", "LPS", "LPS_Dex"), labels=c("Veh", "LPS", "DexLPS"))) genes_OI <- c("Stat1","Stat2","Stat3","Stat4","Stat5a","Stat5b","Stat6","Tcf7","Pou2f1","Rel","Nfkb1","Meis1","Nr3c1") gg_expr <- ggplot(expr_long %>% filter(mgi_symbol %in% genes_OI) %>% mutate(across(mgi_symbol, factor, levels=genes_OI)) )+ geom_abline(slope=0, intercept = 0, colour="grey")+ #ggplot(expr_long %>% filter(mgi_symbol %in% c("Stat2","Stat3","Tcf7","Pou2f1","Rel","Nfkb1","Meis1","Nr3c1")))+ geom_boxplot(aes(x=condition, y=log2(value)))+ geom_point(aes(x=condition, y=log2(value)))+ facet_wrap(~mgi_symbol, ncol=7)+ labs( y="log2 (FPKM)")+ theme(axis.text.x = element_text(angle=30, hjust=1)) gg_expr #------------------------------------------------------------------------------- # TF activities #------------------------------------------------------------------------------- TFA_long <- as.data.frame(TFA) %>% tibble::rownames_to_column( "TF") %>% tidyr::pivot_longer(cols=2:(ncol(TFA)+1), names_to="sample", values_to="tfactivity") %>% tidyr::extract( "sample", c("condition","rep","type"), regex = "(.*)([0-9])_([^_]+)$") %>% mutate(condition=factor(condition, levels=c("V", "LPS", "LPS_Dex"), labels=c("Veh", "LPS", "DexLPS"))) gg_TFA_all <- ggplot(TFA_long) + geom_boxplot(aes(x=condition, y=tfactivity))+ facet_wrap(~TF, scales = "free", ncol=6) ggsave(here("./results/current/tfactivity/TFA_all.png"), gg_TFA_all, width=190, height = 250, units="mm") gg_tfa <- ggplot(TFA_long %>% filter(TF %in% c("NR3C1","STAT2","STAT3","RELA","JUN","JUNB","JUND","FOS","FOSL"))) + geom_boxplot(aes(x=condition, y=tfactivity))+ geom_point(aes(x=condition, y=tfactivity))+ labs( y = "TF activity")+ theme(axis.text.x = element_text(angle=45, hjust=1))+ facet_wrap(~TF, scales = "free") #------------------------------------------------------------------------------- # ChIP-MS data #------------------------------------------------------------------------------- chipMS <- openxlsx::read.xlsx( opt$chipms ) # save original column names to know what is what orig_cnames <- colnames(chipMS) # make them valid for R colnames(chipMS) <- make.names(colnames(chipMS)) #orig_cnames #colnames(chipMS) genes_OI <- c("Rela","Rel","Junb","Nfkb1","Nr3c1","Meis1","Stat1","Stat2","Stat3","Stat4","Stat5b;Stat5a","Stat6") gg_chipMS <- ggplot(chipMS, aes(x=Test.statistic ,y=X.Log.t.test.p.value ))+ geom_point(colour="grey")+ geom_abline(slope=0, intercept = 0, colour="grey")+ geom_point(data=chipMS %>% filter(X.Log.t.test.p.value > 1.3), colour="black")+ geom_point(data=chipMS %>% filter(Gene.names %in% genes_OI), colour="blue", size=2)+ ggrepel::geom_label_repel(data=chipMS %>% filter(Gene.names %in% genes_OI), aes( label=Gene.names ), size=2, nudge_x = 0.5, nudge_y = 0.2)+ geom_hline(yintercept=1.3, linetype=2)+ labs(x="Test statistic - wtGR vs wtIgG", y="-log(p) - wtGR vs wtIgG") gg_chipMS #------------------------------------------------------------------------------- # deeptools at STAT #------------------------------------------------------------------------------- #deeptools <- png::readPNG( opt$heatmap ) #gg_deeptools <- ggplot() + # ggpubr::background_image(deeptools) + # # so it doesn't get squished # #coord_fixed()+ # # This ensures that the image leaves some space at the edges # theme(plot.margin = margin(t=0, l=0, r=0, b=0, unit = "cm"), # axis.line = element_blank()) # --> moved to supplements #------------------------------------------------------------------------------- # STAT motifs #------------------------------------------------------------------------------- gg_statmotifs <- meme_db_expressed %>% filter( grepl("STAT", stringr::str_to_upper(altname)) ) %>% mutate(name = paste(stringr::str_to_upper(altname), "(", name, ")")) %>% to_list() %>% view_motifs() gg_statmotifs ggsave(here("./results/current/Figures/Suppl_Figure_statmotifs.pdf"), gg_statmotifs, width=180, height=200, units="mm", bg="white") #------------------------------------------------------------------------------------ #------ plot aggregated figure #------------------------------------------------------------------------------------ #gg_c1 <- ggpubr::ggarrange(gg_expr, gg_tfa, # labels = c("A","B"), # ncol = 1, nrow = 2, heights = c(1,1)) left <- ggpubr::ggarrange(gg_expr, gg_difffootprint, labels = c("A","C"), ncol = 1, nrow=2, heights = c(1,1)) right <- ggpubr::ggarrange(gg_chipMS, gg_placeholder, labels = c("B","D"), ncol = 1, nrow=2, heights = c(2,1)) full_panel <- ggpubr::ggarrange(left, right, labels = c(NA,NA), ncol = 2, nrow=1, widths = c(1,0.7)) full_panel ggsave(here("./results/current/Figures/Figure_stats.png"), full_panel, width=190, height=120, units="mm", bg="white") ggsave(here("./results/current/Figures/Figure_stats.pdf"), full_panel, width=190, height=120, units="mm", bg="white") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--auc"), type="character", help="Path to rds file with auc results of all models"), make_option(c("--dirname_featurematrizes"), type="character", help="directory that the unscaled featurematrizes were saved in"), make_option(c("--dirname_models"), type="character", help="directory that the trained models were saved in"), make_option(c("--outfig"), type="character", help="Path to output Figure") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(glmnet, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=8, family = "ArialMT", colour="black"), title=element_text(size=10, family="ArialMT", colour="black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text = element_text(size=8, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(8, 'points'), #change legend key size legend.key.height = unit(8, 'points'), #change legend key height legend.key.width = unit(8, 'points'), #change legend key width legend.text = element_text(size=8, family="ArialMT", colour="black") ) # compare ROC curves for statistical significance # which ones? # let's start with the best performing one and our reference model # what is the name of the best performing model? AUC_metrics <- readRDS ( opt$auc ) bestmodel <- AUC_metrics %>% filter(net_test == max(AUC_metrics$net_test)) %>% rownames() AUC_metrics[bestmodel,"net_train"] # what is the performance of the best model on the training set? #----------------------------------------- # 0 write function that takes model and featurematrix as input and returns ROC object as output not_all_na <- function(x) any(!is.na(x)) get_roc_object <- function(path_performance_rds, path_featurematrix_unscaled_rds){ # 1. load cv.fit of model performance<- readRDS( path_performance_rds ) cvfit_net <- performance[[4]] # 2. load testset of model motifdata <- readRDS( path_featurematrix_unscaled_rds ) motifdata_scaled <- motifdata %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector)))%>% dplyr::select(where(not_all_na)) motifdata_tranval_idx <- motifdata_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) features_train <- motifdata_scaled[ motifdata_tranval_idx, -c(1,2,3)] %>% as.matrix() features_test <- motifdata_scaled[ -motifdata_tranval_idx, -c(1,2,3)] %>% as.matrix() targets_train <- motifdata_scaled[ motifdata_tranval_idx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() targets_test <- motifdata_scaled[ -motifdata_tranval_idx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() # 3. Predict and return ROC targets_net.prob <- predict(cvfit_net, type="response", newx = features_test, s = 'lambda.min') roc_object <- pROC::roc(as.factor(targets_test),targets_net.prob[,1], direction = "<") return(roc_object) } roc_reference <- get_roc_object ( paste0( opt$dirname_models, "prox.rds"), paste0( opt$dirname_featurematrizes, "prox.rds") ) roc_bestmodel <- get_roc_object ( paste0( opt$dirname_models, bestmodel,".rds"), paste0( opt$dirname_featurematrizes, bestmodel,".rds") ) # 4. Compare for pROC # Should we compute the p value comparing our reference to all others and see how many others it beats? #res <- pROC::roc.test( roc_reference, roc_bestmodel, method="delong", alternative="greater") #res #------------------------------------------------------------------------------- # comparison with reference model (proximity based) #------------------------------------------------------------------------------- # initialize a dataframe that will save the name of the model, the pvalue and whether the referene was better all_pairwise_wprox <- data.frame( pvalues=numeric()) for (modelname in rownames(AUC_metrics) ) { # skip the prox model since this is what we compare to if (modelname=="prox"){next} print(modelname) roc_model <- get_roc_object ( paste0( opt$dirname_models, modelname,".rds"), paste0( opt$dirname_featurematrizes, modelname,".rds") ) roc_test_res <- pROC::roc.test( roc_reference, roc_model, method="delong", alternative="greater") new_metrics <- data.frame( p.values=roc_test_res$p.value) all_pairwise_wprox <- rbind(all_pairwise_wprox, new_metrics ) %>% magrittr::set_rownames(c(rownames(all_pairwise_wprox), modelname)) } all_pairwise_wprox$input_is_summitregion <- grepl("motifcounts_summitregion*",rownames(all_pairwise_wprox)) storey_wprox <- qvalue::qvalue(all_pairwise_wprox$p.values, lambda=seq(0.05, 0.8, 0.05), fdr.level = 0.05) storey_wprox print("Pi0 computed by Storey's method for comparison with reference model:") print(storey_wprox$pi0) # plot all p values to estimate where the histogram "levels out" and set this as lambda in Storey's q value gg_storey_wprox <- ggplot(all_pairwise_wprox)+ geom_histogram(aes(p.values,group=input_is_summitregion,fill=input_is_summitregion), bins=40 )+ scale_fill_manual("", labels = c("active regions", "GR summitregions"), breaks = c("FALSE", "TRUE"), values = c("chocolate4", "purple")) + scale_colour_manual("", labels = c("active regions", "GR summitregions"), breaks = c("FALSE", "TRUE"), values = c("chocolate4", "purple")) + labs(x="p value", y="count")+ theme(legend.position = c(0.7, 0.8)) gg_storey_wprox table(all_pairwise_wprox$p.values<=0.05, all_pairwise_wprox$input_is_summitregion) table(storey_wprox$significant) #------------------------------------------------------------------------------- # comparison with best performing model #------------------------------------------------------------------------------- # initialize a dataframe that will save the name of the model, the pvalue and whether the referene was better all_pairwise_wbest <- data.frame(pvalues=numeric()) for (modelname in rownames(AUC_metrics) ) { # skip the bestmodel since this is what we compare to if (modelname==bestmodel){ print("Skipping self") next} print(modelname) roc_model <- get_roc_object ( paste0(opt$dirname_models, modelname,".rds"), paste0( opt$dirname_featurematrizes, modelname,".rds") ) roc_test_res <- pROC::roc.test( roc_bestmodel, roc_model, method="delong", alternative="greater") new_metrics <- data.frame(p.values=roc_test_res$p.value) all_pairwise_wbest <- rbind(all_pairwise_wbest, new_metrics ) %>% magrittr::set_rownames(c(rownames(all_pairwise_wbest), modelname)) } all_pairwise_wbest$input_is_summitregion <- grepl("motifcounts_summitregion*",rownames(all_pairwise_wbest)) storey_wbest <- qvalue::qvalue(all_pairwise_wbest$p.values, lambda=seq(0.05, 0.35, 0.05)) storey_wbest print("Pi0 computed by Storey's method for comparison with best model:") print(storey_wbest$pi0) # plot all p values to estimate where the histogram "levels out" and set this as lambda in Storey's q value gg_storey_wbest <- ggplot(all_pairwise_wbest)+ geom_histogram(aes(p.values,group=input_is_summitregion,fill=input_is_summitregion), bins=40 )+ scale_fill_manual("", labels = c("active regions", "GR summitregions"), breaks = c("FALSE", "TRUE"), values = c("chocolate4", "purple")) + scale_colour_manual("", labels = c("active regions", "GR summitregions"), breaks = c("FALSE", "TRUE"), values = c("chocolate4", "purple")) + labs(x="p value", y="count")+ theme(legend.position = c(0.7, 0.8)) gg_storey_wbest #------------------------------------------------------------------------------- # bivariate models with motifs of interest #------------------------------------------------------------------------------- # Load in featurematrix, scale it and fit model with out motifs of interest in bivariate model motifdata <- readRDS( paste0(opt$dirname_featurematrizes, "prox.rds" )) motifdata_scaled <- motifdata %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) motifdata_tranval_idx <- motifdata_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) features_train <- motifdata_scaled[ motifdata_tranval_idx, -c(1,2,3)] %>% as.matrix() features_test <- motifdata_scaled[ -motifdata_tranval_idx, -c(1,2,3)] %>% as.matrix() targets_train <- motifdata_scaled[ motifdata_tranval_idx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() targets_test <- motifdata_scaled[ -motifdata_tranval_idx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() model_coefficients <- data.frame(motifname = character(), intercept = numeric(), coefficient = numeric(), pvalue = numeric() ) for (motif in c("MEIS1","NFKB1","REL","POU2F1","TCF7","STAT3")){ glm_res <- glm(targets_train ~ features_train[,c(motif)], family="binomial") new_model <- data.frame(motifname=motif, intercept= glm_res$coefficients[1], coefficient = glm_res$coefficients[2], pvalue = coef(summary(glm_res))[2,4], row.names = NULL) model_coefficients <- rbind( model_coefficients, new_model ) } gg_bivariate <- ggplot(data=model_coefficients)+ geom_point(aes(coefficient, -log10(pvalue)))+ ggrepel::geom_label_repel(aes(coefficient, -log10(pvalue), label=motifname), size=2)+ coord_cartesian(ylim=c(0,max(-log10(model_coefficients$pvalue))), xlim=c(min(model_coefficients$coefficient),-0.1))+ geom_hline(aes(yintercept=-log10(0.05)), linetype="dashed" ) #------------------------------------------------------------------------------- # model performance on training set #------------------------------------------------------------------------------- melted_AUC_metrics <- AUC_metrics %>% reshape::melt(., measure.vars=c("net_train","net_test")) #remove the proximity based model and add it in shape of a reference line instead prox_reference_train <- melted_AUC_metrics %>% filter(variable=="net_train") %>% filter(onlymax=="prox") %>% pull(value) %>% as.numeric() gg_AUC_train <- ggplot(melted_AUC_metrics %>% filter(variable=="net_train") %>% filter(onlymax!="prox"))+ geom_bar(aes(x=weight, y=value, fill=condition, alpha=onlymax), colour="black", size=0.5, width=0.9, stat="identity", position=position_dodge())+ geom_hline(aes(yintercept=prox_reference_train, linetype="reference"), colour="purple")+ scale_linetype_manual(name="", values = c(2))+ facet_grid( cols = vars(sepPromEnh,excludepromoters), rows= vars(motifdata), labeller = labeller( motifdata=c("motifcounts_abcregion"="active regions", "motifcounts_summitregion"="GR summitregions"), excludepromoters=c("FALSE"="excl.prom: none", "onlyNONself"="excl.prom: nonself", "all"="excl.prom: all"), sepPromEnh=c("TRUE"="sep prom and enh", "FALSE"= "aggr prom and enh") ) )+ scale_fill_manual( values=c("#339966","#0066CC", "orange"), breaks=c("dexlps","lps", "DexLPS-LPS"), labels=c("DexLPS","LPS", "\u0394 DexLPS-LPS") )+ scale_alpha_manual( values=c(1,0.5), breaks=c("FALSE","TRUE"), labels=c("1-to-many", "1-to-1") )+ labs(x="Weight features during aggregation", y="AUC", alpha="mapping" )+ theme( legend.position = "bottom") #------------------------------------------------------------------------------- # arrange and save figure #------------------------------------------------------------------------------- gg_placeholder <- ggplot() + theme(plot.margin = margin(t=0, l=0, r=0, b=0, unit = "cm"), axis.line = element_blank()) #A in this panel will be tha training performance gg_c1 <- ggpubr::ggarrange(gg_storey_wbest, gg_storey_wprox, labels = c("B","C"), ncol = 1, nrow = 2) gg_c2 <- ggpubr::ggarrange(gg_bivariate, gg_placeholder, labels = c("D",NA), ncol = 1, nrow = 2, heights = c(1.5,1)) gg_r2 <- ggpubr::ggarrange(gg_c1, gg_c2, labels = c(NA, NA), ncol = 2, nrow = 1) gg <- ggpubr::ggarrange(gg_AUC_train, gg_r2, labels = c("A", NA), ncol = 1, nrow = 2,heights = c(1,1)) gg ggsave( opt$outfig, gg, width=190, height=200, units="mm", bg="white") # also save it as pdf ggsave( gsub(".png",".pdf",opt$outfig), gg, width=190, height=200, units="mm", bg="white") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--fimo"), type="character", help="Path to fimo rds file that should be subsetted"), make_option(c("--summit_granges"), type="character", help="Path to summit granges file that we use to narrow down fimo hits on summitregions"), make_option(c("--motif_altname"), type="character", help="Altname of motif that we filter the fimo results for (case sensitive!)"), make_option(c("--outfile"), type="character", help="Path to subsetted output bed file") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(rtracklayer, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(GenomicRanges, warn.conflicts=F, quietly=T)) fimo_results <- readRDS ( opt$fimo ) ChIPseq_summit_Granges <- readRDS( opt$summit_granges ) # Take 100bp windows around ChIP-seq summits summit_flank_100bp <- ChIPseq_summit_Granges %>% plyranges::anchor_center() %>% plyranges::mutate(width = 100) # narrow the hits of the motif of interest down to the immediate summit region motifhits <- fimo_results %>% filter(motif_alt_id==opt$motif_altname) %>% filter_by_overlaps(summit_flank_100bp) print(paste0("We found " , length(motifhits), " motifhits for ", opt$motif_altname)) export.bed(motifhits, opt$outfile ) |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("-g", "--genekey"), type="character", help="Path to biomart derived genekey"), make_option(c("-e", "--expression"), type="character", help="Path to TPM expression matrix") ) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) opt <- parse_args(OptionParser(option_list=option_list)) tpm_counts <- as.matrix(read.table( opt$expression, quote = "\"", header=TRUE, row.names = 1 )) geneKey_GRCm38.p6 <- read.table( opt$genekey, header=TRUE, sep="\t") tpm_counts_ext <- merge(as.data.frame(tpm_counts), geneKey_GRCm38.p6, by.x="row.names", by.y="ensembl_gene_id", all.x=TRUE) # compute the mean per condition tpm_counts_ext <- tpm_counts_ext %>% rowwise() %>% mutate(mean_dexlps=mean(c(LPS_Dex1_nascent, LPS_Dex2_nascent, LPS_Dex3_nascent))) %>% mutate(mean_lps=mean(c(LPS1_nascent, LPS2_nascent, LPS3_nascent ))) %>% mutate(mean_veh=mean(c(V1_nascent, V2_nascent ))) write.table(tpm_counts_ext[,c("Row.names","mean_veh")], file="results/current/abcmodel/expression/Veh_tpm.tsv", quote = FALSE, row.names = FALSE, col.names = FALSE, sep="\t") write.table(tpm_counts_ext[,c("Row.names","mean_lps")], file="results/current/abcmodel/expression/LPS_tpm.tsv", quote = FALSE, row.names = FALSE, col.names = FALSE, sep="\t") write.table(tpm_counts_ext[,c("Row.names","mean_dexlps")], file="results/current/abcmodel/expression/DexLPS_tpm.tsv", quote = FALSE, row.names = FALSE, col.names = FALSE, sep="\t") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--ctss_pool1"), type="character", help="Path for ctss file from FANTOM5 of pool 1"), make_option(c("--ctss_pool2"), type="character", help="Path for ctss file from FANTOM5 of pool 2"), make_option(c("--liftoverchain"), type="character", help="Path to liftover chain file for mm9 to mm10"), make_option(c("--gencode_mm9_geneanno"), type="character", help="Path to genomic reference file for mm9"), make_option(c("--gencode_mm10_geneanno"), type="character", help="GENCODE genomic reference for assembly mm10, prefiltered for gene entries"), make_option(c("--outdir"), type="character", help="Output directory") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(BSgenome.Mmusculus.UCSC.mm9, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ChIPseeker, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(CAGEr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(rtracklayer, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(GenomicRanges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(org.Mm.eg.db, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(TxDb.Mmusculus.UCSC.mm10.knownGene, warn.conflicts=F, quietly=T)) outdir <- here(opt$outdir) # Workflow is based on the CAGEr vignette # https://www.bioconductor.org/packages/release/bioc/vignettes/CAGEr/inst/doc/CAGEexp.html #---------------------- Import CAGE samples from BMDMs #------------------------------------------------------------------------------------ # get URL for public samples, download and read necessariy columns into ctss file (see snakemake workflow) # We want to import bone-marrow derived macrophage samples through CAGEr. # After looking at list of available datasets we can decide what samples best fit our needs. # Let's see what samples are available through FANTOM5 #data(FANTOM5mouseSamples) #head(FANTOM5mouseSamples) # The FANTOM5 dataframe holds descriptions of the samples and the url where they can be retrieved. # There's an easy way to import samples that match a certain term into a CAGEset object. #mac_samples <- FANTOM5mouseSamples[grep("macrophage, bone marrow derived", # FANTOM5mouseSamples[,"description"]),] print("NOTE: reference genome for the public CAGE files is mm9!") ce <- CAGEr::CAGEexp( genomeName = "BSgenome.Mmusculus.UCSC.mm9", inputFiles = c(opt$ctss_pool1 ,opt$ctss_pool2), inputFilesType = "ctss", sampleLabels = c("pool1","pool2") ) # To actually read in the data into the object we use getCTSS() function, that will add an experiment called tagCountMatrix to the CAGEexp object. ce <- CAGEr::getCTSS(ce) ce #------------------------------------------------------------------------------------ #---------------------- QC #------------------------------------------------------------------------------------ ncbim37_anno <- rtracklayer::import.gff(opt$gencode_mm9_geneanno) ce <- annotateCTSS(ce, ncbim37_anno) colData(ce)[,c("librarySizes", "promoter", "exon", "intron", "unknown")] plotAnnot(ce, "counts") corr.m <- plotCorrelation2( ce, samples = "all" , tagCountThreshold = 1, applyThresholdBoth = FALSE , method = "pearson") #------------------------------------------------------------------------------------ #---------------------- Get read clusters #------------------------------------------------------------------------------------ print("Merging samples") #Now we can merge them ce <- mergeSamples(ce, mergeIndex = c(1,1), mergedSampleLabels = c("BMDM")) # redo annotation since this gets reset during merging ce <- annotateCTSS(ce, ncbim37_anno) print("The total library size is:") print(librarySizes(ce)) # Check if data follows a power law distribution plotReverseCumulatives(ce, fitInRange = c(5, 3000), onePlot = TRUE) print("Normalizing reads") # Since we don't really care about making comparisons between different population we could prob just skip the normalization # The fit range is chosen from the plot. We take the alpha from the ref distribution and set T to a million to get the tag count per million (TPM) ce <- normalizeTagCount(ce, method = "powerLaw", fitInRange = c(5, 3000), alpha = 1.15, T = 1*10^6) #mac_CAGEset@tagCountMatrix print("Cluster the tags") # After normalization we can cluster the tags. # Clustering, only seems to work with the CAGEset object (due to some problems with the IRanges column) # From the CAGEr vignette: # "Transcription start sites are found in the promoter region of a gene and reflect the transcriptional activity of that promoter (Figure 5). TSSs in the close proximity of each other give rise to a functionally equivalent set of transcripts and are likely regulated by the same promoter elements. Thus, TSSs can be spatially clustered into larger transcriptional units, called tag clusters (TCs) that correspond to individual promoters. CAGEr supports three methods for spatial clustering of TSSs along the genome, two ab initio methods driven by the data itself, as well as assigning TSSs to predefined genomic regions:" ce <- clusterCTSS(ce, threshold=1, thresholdIsTpm = TRUE, nrPassThreshold = 1, method="distclu", maxDist=20, removeSingletons = TRUE, keepSingletonsAbove = 3) # Let's have a look what the result looks like head(tagClustersGR(ce, sample = "BMDM")) # calculate cumulative distribution for every tag cluster in each of the samples ce <- cumulativeCTSSdistribution(ce, clusters = "tagClusters", useMulticore = T) # determine the positions of selected quantiles ce <- quantilePositions(ce, clusters = "tagClusters", qLow = 0.1, qUp = 0.9) # How many tagclusters do we have in total? length(tagClustersGR(ce, sample = "BMDM")) # histogram of interquantile width plotInterquantileWidth(ce, clusters = "tagClusters", tpmThreshold = 3, qLow = 0.1, qUp = 0.9) print("Retrieving clusters as GenomicRanges") clusters_gr <- tagClustersGR(ce, sample="BMDM") #------------------------------------------------------------------------------------ #---------------------- Liftover coordinates to mm10 #------------------------------------------------------------------------------------ print("Liftover to mm10") # * Now we can lift over the intervals to mm10 # * Annotate them with peakanno # * pick the most highly expressed one for each gene liftover <- function(peaks_gr_mm9){ #input is a GenomicRanges object in mm9 coordinates #lift peak locations from mm9 to mm10 chain <- rtracklayer::import.chain(opt$liftoverchain) on.exit( close( file(opt$liftoverchain)) ) peaks_gr_mm10 <- rtracklayer::liftOver(peaks_gr_mm9, chain) peaks_gr_mm10 <- GenomicRanges::GRanges(unlist(peaks_gr_mm10)) return(peaks_gr_mm10) } mac_cage_mm10 <- liftover( clusters_gr ) ggplot(as.data.frame(mac_cage_mm10), aes(x=width)) + geom_histogram(bins = 100) # Liftover coordinates of dominant_ctss dominant_ctss <- liftover( GRanges( seqnames = seqnames(clusters_gr), ranges = IRanges(start = clusters_gr$dominant_ctss, end = clusters_gr$dominant_ctss), score=clusters_gr$score) ) #------------------------------------------------------------------------------------ #----------------- Annotate TSS clusters to reference gene coordinates #------------------------------------------------------------------------------------ print("Annotate TSS coordinates") # Use coordinates of the dominant ctss downstream mac_cage_anno <- ChIPseeker::annotatePeak(dominant_ctss, tssRegion=c(-1000, 1000), #more stringent than default level = "gene", TxDb=TxDb.Mmusculus.UCSC.mm10.knownGene, annoDb = "org.Mm.eg.db") # For those that are reasonably close to a TSS, # check for each gene, which position has the highest score. mac_cage_maxscore <- as.data.frame(mac_cage_anno) %>% filter(abs(distanceToTSS)<=30000) %>% mutate(SYMBOL=as.factor(SYMBOL)) %>% filter(SYMBOL!="") %>% group_by(SYMBOL)%>% filter(score == max(score))%>% filter(distanceToTSS == min(distanceToTSS )) # for tied score, use shorter distance nrow(mac_cage_maxscore) ggplot(mac_cage_maxscore, aes(x=distanceToTSS)) + geom_histogram(bins = 100) #------------------------------------------------------------------------------------ #---------- retrieve gene coordinates and promoterregion from reference #------------------------------------------------------------------------------------ gencode_mm10_geneanno <- rtracklayer::import.gff(opt$gencode_mm10_geneanno) genecoords <- as.data.frame(gencode_mm10_geneanno) %>% dplyr::select("seqnames","start","end","strand","gene_id") %>% mutate(score=0) %>% dplyr::mutate(gene_id=gsub("\\.[0-9]*$","",gene_id)) %>% dplyr::filter(!"gene_id"=="")%>% dplyr::select("seqnames","start","end","gene_id","score","strand") gencode_mm10_promoterregions <- promoters(gencode_mm10_geneanno) gencode_mm10_promoterregions <- as.data.frame(gencode_mm10_promoterregions) %>% dplyr::select("seqnames","start","end","strand","gene_id") %>% mutate(score=0) %>% dplyr::mutate(gene_id=gsub("\\.[0-9]*$","",gene_id)) %>% dplyr::filter(!"gene_id"=="")%>% dplyr::select("seqnames","start","end","gene_id","score","strand") #------------------------------------------------------------------------------------ #----------------- export files #------------------------------------------------------------------------------------ write.table(genecoords, file = paste0(outdir,"reference_genecoords.bed"), sep="\t", col.names = FALSE, quote=FALSE, row.names = FALSE) write.table(gencode_mm10_promoterregions, file = paste0(outdir,"reference_promoterregions.bed"), sep="\t", col.names = FALSE, quote=FALSE, row.names = FALSE) rtracklayer::export.bed(as.data.frame(mac_cage_mm10), paste0(outdir, "mac_cage_tssclusterregions.bed"), format="bed") rtracklayer::export.bed(as.data.frame(dominant_ctss), paste0(outdir, "mac_cage_dominant_ctss.bed"), format="bed") rtracklayer::export.bed(mac_cage_maxscore, paste0(outdir,"mac_cage_maxscore.bed"), format="bed") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--summit_granges"), type="character", help="Path to rds file of summits in granges format with directionschange and distancetoTSS as additional metadata columns"), make_option(c("--memedb_expressed"), type="character", help="Path to memedb file filtered for motifs where TFs are expressed in 4sU")) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(memes, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(universalmotif, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(GenomicRanges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(BSgenome.Mmusculus.UCSC.mm10, warn.conflicts=F, quietly=T)) #------------------------------- ## Import reference for sequence #------------------------------- mm.genome <- BSgenome.Mmusculus.UCSC.mm10 #------------------------------- ## read in prepared data #------------------------------- ChIPseq_summit_Granges <- readRDS(opt$summit_granges) # Take 100bp windows around ChIP-seq summits summit_flank_100bp <- ChIPseq_summit_Granges %>% plyranges::anchor_center() %>% plyranges::mutate(width = 100) # Take 100bp windows around ChIP-seq summits summit_flank_1000bp <- ChIPseq_summit_Granges %>% plyranges::anchor_center() %>% plyranges::mutate(width = 1000) meme_db_expressed <- readRDS(opt$memedb_expressed) # to_list() converts the database back from data.frame format to a standard `universalmotif` object. options(meme_db = to_list(meme_db_expressed, extrainfo = FALSE)) # where is meme installed my_memepath="~/software/meme/bin/" check_meme_install(meme_path=my_memepath) summit_flank_100bp_seq <- summit_flank_100bp %>% get_sequence(mm.genome) summit_flank_1000bp_seq <- summit_flank_1000bp %>% get_sequence(mm.genome) #------------------------------- ## run fimo #------------------------------- fimo_results <- runFimo(summit_flank_1000bp_seq, meme_db_expressed, meme_path=my_memepath) saveRDS (fimo_results, here("results/current/memes_bioc/fimo_1000bp/fimo.rds") ) print("Finished running fimo") print("Analysis DONE") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--ABC_all"), type="character", help="path to abc results of dexlps condition"), make_option(c("--memedb_expressed"), type="character", help="Path to memedb file filtered for motifs where TFs are expressed in 4sU"), make_option(c("--output"), type="character", help="fimo results file") ) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(memes, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(universalmotif, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(GenomicRanges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(BSgenome.Mmusculus.UCSC.mm10, warn.conflicts=F, quietly=T)) #------------------------------- ## Import reference for sequence #------------------------------- mm.genome <- BSgenome.Mmusculus.UCSC.mm10 #------------------------------- ## read in prepared data #------------------------------- ABC_all <- read.delim(opt$ABC_all) %>% plyranges::as_granges(., seqnames=chr) # no need to run fimo a bunch of times on the same enhancers regions, just because they are listed more than once (with different ABCscores) ABC_unique <- unique(ABC_all) meme_db_expressed <- readRDS(opt$memedb_expressed) # to_list() converts the database back from data.frame format to a standard `universalmotif` object. options(meme_db = to_list(meme_db_expressed, extrainfo = FALSE)) # where is meme installed my_memepath="~/software/meme/bin/" check_meme_install(meme_path=my_memepath) #------------------------------- ## get sequences #------------------------------- enhancer_seq <- ABC_unique %>% get_sequence(mm.genome) #------------------------------- ## run fimo #------------------------------- # conda activate py_3 # perlbrew use perl-5.34.0 # nohup Rscript memes_runanalyses_ABCenhancerregions.r & (from within the script directory) fimo_results <- runFimo(enhancer_seq, meme_db_expressed, meme_path=my_memepath) print("Finished running fimo") saveRDS (fimo_results, opt$output) print("Done saving results") |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--summit_granges"), type="character", help="Path to rds file of summits in granges format with directionschange and distancetoTSS as additional metadata columns"), make_option(c("--memedb_expressed"), type="character", help="Path to memedb file filtered for motifs where TFs are expressed in 4sU")) opt <- parse_args(OptionParser(option_list=option_list)) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(memes, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(universalmotif, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(GenomicRanges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(BSgenome.Mmusculus.UCSC.mm10, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) #------------------------------- ## Import reference for sequence #------------------------------- mm.genome <- BSgenome.Mmusculus.UCSC.mm10 #------------------------------- ## read in prepared data #------------------------------- ChIPseq_summit_Granges <- readRDS(opt$summit_granges) # Take 100bp windows around ChIP-seq summits summit_flank_100bp <- ChIPseq_summit_Granges %>% plyranges::anchor_center() %>% plyranges::mutate(width = 100) # Take 100bp windows around ChIP-seq summits summit_flank_1000bp <- ChIPseq_summit_Granges %>% plyranges::anchor_center() %>% plyranges::mutate(width = 1000) meme_db_expressed <- readRDS(opt$memedb_expressed) # to_list() converts the database back from data.frame format to a standard `universalmotif` object. options(meme_db = to_list(meme_db_expressed, extrainfo = FALSE)) # where is meme installed my_memepath="~/software/meme/bin/" check_meme_install(meme_path=my_memepath) #------------------------------- # define inputs #------------------------------- summit_flank_100bp_seq <- summit_flank_100bp %>% get_sequence(mm.genome) summit_flank_1000bp_seq <- summit_flank_1000bp %>% get_sequence(mm.genome) summit_flank_seq_bydirchange <- summit_flank_100bp %>% # remove unchanged ones and only compare "up" vs "down" filter(directionchange !="ns")%>% # Get a list of chip peaks belonging to each set split(mcols(.)$directionchange) %>% # look up the DNA sequence of each peak within each group get_sequence(mm.genome) #------------------------------- ## up vs downregulation # run by directionchange to discover consensus motif separately #------------------------------- #------------------------------- # STREME #------------------------------- print("Start running streme for 100bp") stremeout_100bp_down <- here("results/current/memes_bioc/streme_100bp_down/streme.xml") if (!file.exists( stremeout_100bp_down )){ runStreme(summit_flank_seq_bydirchange[["down"]], control="shuffle", objfun="de", meme_path="~/software/meme/bin/", silent=FALSE, outdir = dirname(stremeout_100bp_down)) } stremeout_100bp_up <- here("results/current/memes_bioc/streme_100bp_up/streme.xml") if (!file.exists( stremeout_100bp_up )){ runStreme(summit_flank_seq_bydirchange[["up"]], control="shuffle", objfun="de", meme_path="~/software/meme/bin/", outdir = dirname(stremeout_100bp_up)) } print("Finished running streme for up- and down-regions") #------------------------------- # DREME #------------------------------- dremeout_100bp_down <- here("results/current/memes_bioc/dreme_100bp_down/dreme.xml") if (!file.exists(dremeout_100bp_down)){ runDreme(summit_flank_seq_bydirchange[["down"]], "shuffle", meme_path="~/software/meme/bin/", outdir = dirname(dremeout_100bp_down)) } dremeout_100bp_up <- here("results/current/memes_bioc/dreme_100bp_up/dreme.xml") if (!file.exists(dremeout_100bp_up)){ runDreme(summit_flank_seq_bydirchange[["up"]], "shuffle", meme_path="~/software/meme/bin/", outdir = dirname(dremeout_100bp_up)) } print("Done running DREME") #------------------------------- ## run ame - discriminative mode #------------------------------- # enriched in upregulated with "down" as control ame_discr_up <- here("results/current/memes_bioc/ame_discr_up/ame.tsv") if (!file.exists( ame_discr_up )){ runAme(summit_flank_seq_bydirchange, control = "down", meme_path=my_memepath, outdir=dirname(ame_discr_up)) } # enriched in downregulated with "up" as control ame_discr_down <- here("results/current/memes_bioc/ame_discr_down/ame.tsv") if (!file.exists(ame_discr_down)){ runAme(summit_flank_seq_bydirchange, control = "up", meme_path=my_memepath, outdir=dirname(ame_discr_down)) } print("Finished running ame in discriminative mode for direction of expressionchange") #------------------------------- ## all summits #------------------------------- ## run streme to discover consensus motif #------------------------------- #print("Starting streme 1000bp") #stremeout_1000bp <- here("results/current/memes_bioc/streme_1000bp/streme.xml") #if (!file.exists( stremeout_1000bp )){ # runStreme(summit_flank_1000bp_seq, control="shuffle", # meme_path="~/software/meme/bin/", # outdir = dirname(stremeout_1000bp) ) #} print("Starting streme 100bp for all summits") stremeout_100bp <- here("results/current/memes_bioc/streme_100bp/streme.xml") if (!file.exists( stremeout_100bp )){ runStreme(summit_flank_100bp_seq, control="shuffle", meme_path="~/software/meme/bin/", outdir = dirname(stremeout_100bp) ) } print("Finished running streme for 100bp summit regions") # Option objfun="cd" does not seem to get passed on to streme #print("Starting streme 1000bp with central enrichment") #stremeout_cd_1000bp <- here("results/current/memes_bioc/streme_cd_1000bp/streme.xml") #if (!file.exists( stremeout_cd_1000bp )){ # runStreme(summit_flank_1000bp_seq[1:100], objfun="cd", control=NA, # meme_path="~/software/meme/bin/", # outdir = dirname(stremeout_cd_1000bp) ) #} #print("Finished running streme for 1000bp summit regions") print("Analysis DONE") |
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c( "--contrast_DexLPSvLPS"), type="character", help="Path to annotated tsv file of DeSeq2 contrast of DexLPS vs LPS"), make_option(c("--assignment_summit_prox"), type="character", help="Path to rds file of proximity based assignment of peak summits to genes"), make_option(c("--assignment_summits_abcregion_dexlps"), type="character", help="Path to rds file of assignment of peak summits within abcregions to genes (in DexLPS condition)"), make_option(c("--assignment_summits_abcregion_lps"), type="character", help="Path to rds file of assignment of peak summits within abcregions to genes (in LPS condition)"), make_option(c("--assignment_abcregion_dexlps"), type="character", help="Path to rds file of assignment of abcregions to genes (in DexLPS condition)"), make_option(c( "--assignment_abcregion_lps"), type="character", help="Path to rds file of assignment of abcregions to genes (in LPS condition)"), make_option(c("--motifcounts_summitregion"), type="character", help="Path to rds file of fimo motifcounts within summitregions"), make_option(c("--motifcounts_abcregion_dexlps"), type="character", help="Path to rds file of fimo motifcounts within ABC regions (in DexLPS condition)"), make_option(c("--motifcounts_abcregion_lps"), type="character", help="Path to rds file of fimo motifcounts within ABC regions (in LPS condition)"), make_option(c("--model_coefs_joint"), type="character", help="Path to rds file with model coefficients of joint models"), make_option(c("--model_coefs_sep"), type="character", help="Path to rds file with model coefficients of models tha include enhancers and promoters separately"), make_option(c( "--featuredir"), type="character", help="Path to directory with unscales featurematrizes"), make_option(c( "--outfile"), type="character", help="Path to output file with metrics") ) opt <- parse_args(OptionParser(option_list=option_list)) dir.create(opt$featuredir) #change default for stringAsFactors options(stringsAsFactors = FALSE) suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plyranges, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(org.Mm.eg.db, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(DESeq2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ComplexHeatmap, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) #set defaults for ggplot2 figures theme_update(panel.background = element_rect(fill = "transparent", colour = NA), plot.background = element_rect(fill = "transparent", colour = NA), legend.background = element_rect(fill = "transparent", colour = NA), legend.key = element_rect(fill = "transparent", colour = NA), text=element_text(size=6, family = "ArialMT", colour="black"), title=element_text(size=8, family="ArialMT", colour="black"), panel.grid.major = element_line(colour="grey", size=0.2), panel.grid.minor = element_blank(), axis.text = element_text(size=6, family="ArialMT", colour="black"), axis.line = element_line(colour="black"), axis.ticks = element_line(colour="black"), legend.key.size = unit(6, 'points'), #change legend key size legend.key.height = unit(6, 'points'), #change legend key height legend.key.width = unit(6, 'points'), #change legend key width legend.text = element_text(size=6, family="ArialMT", colour="black")) set.seed(12345) #------------------------------- ## read in data #------------------------------- contrast_DexLPSvLPS <- read.delim(opt$contrast_DexLPSvLPS) for (optname in names(opt)[2:11]){ #except for the featuredir print(paste0("Loading ", optname)) assign(optname, readRDS( opt[[optname]] )) } #-------------------------------------------- ## ---- function definitions #-------------------------------------------- coerce_coef2df <- function(model_coef){ model_coef <- as.matrix(model_coef) model_coef <- as.data.frame(model_coef) model_coef$names <- rownames(model_coef) colnames(model_coef) <- c("estimates", "names") return(model_coef) } #------------function will------------: # * merge the motifdata with gene assignments and # * then merge the gene expression changes, # * filter for DE genes and aggregate per gene #-------------------------------------- merge_motifdata_with_assignments <- function( motifcounts, assignments, contrast, maxonly=FALSE, excludepromoters=FALSE, weightby=FALSE, sepPromEnh=FALSE){ # check if we should use all abc assignments, or only the max one of each peakID if(maxonly==TRUE){ assignments <- assignments %>% group_by(name) %>% filter(abcscore==max(abcscore)) %>% distinct() } else{ assignments <- assignments } if(excludepromoters=="all"){ assignments <- assignments %>% filter(!class=="promoter") } else if (excludepromoters=="onlyNONself"){ assignments <- assignments %>% filter(!c(class=="promoter" & isSelfPromoter=="False")) } else{ assignments <- assignments } motifdf <- merge(motifcounts, assignments, by.x="name", by.y="name") motifdf <- motifdf %>% relocate(c(anno)) # merge the expression change # use mgi_symbol for prox based, otherwise ensemblID # We DONT set all.x=TRUE because we don't care about predicting gene that aren't even expressed or that we don't have a clear label for if(maxonly=="prox"){ motifdf <- merge(motifdf, contrast, by.x="anno", by.y="mgi_symbol") } else { motifdf <- merge(motifdf, contrast, by.x="anno", by.y="Row.names") } #recode the logFC and padj into a label (optionally through command line arguments) motifdf <- motifdf %>% mutate(label=case_when(log2FoldChange>0.58 & padj < 0.05 ~ "up", log2FoldChange<(-0.58) & padj < 0.05 ~ "down", TRUE ~ "no_change")) %>% filter(label!="no_change") %>% mutate(label=factor(label, levels=c("down","up"), labels=c(0,1))) %>% relocate(label) # chr 2, 3 and 4 (20%) were used as the tuning set for hyperparameter tuning. # Regions from chromosomes 1, 8 and 9 (20%) were used as the test set for performance evaluation # The remaining regions were used for model training. #------aggregate over gene SYMBOL if ("abcscore" %in% colnames(motifdf)) { # loop for ABC based assignments if (weightby=="abcscore"){ unselect_col <- "abcnumerator" } else if(weightby=="abcnumerator") { unselect_col <- "abcscore" } else{ unselect_col <- c("abcscore","abcnumerator") } if (sepPromEnh==TRUE){ motifdf_aggr <- motifdf %>% dplyr::select(!c(unselect_col,"name","baseMean","log2FoldChange","lfcSE","stat","pvalue","padj","gene_biotype","mgi_symbol")) %>% { if(weightby!=FALSE) mutate(.,across(where(is.numeric), ~ (.x * get(weightby)))) else .} %>% # weight features by score dplyr::select(!any_of(as.character(weightby))) %>% # then we can drop the score since it will be nonsensical after the aggregation anyways group_by(label,seqnames,anno,class) %>% dplyr::summarise(across( where(is.numeric), .fns=sum )) %>% # sum up genewise feature counts ungroup() # From here we need to cast the motifcounts for the promoterregions, so that in the end we have one row per gene (instead of 1-2) motifdf_aggr <- motifdf_aggr %>% tidyr::pivot_wider(id_cols=c(label,seqnames,anno), names_from=class, values_from = !c(label,seqnames,anno, class), values_fill = 0) } else { motifdf_aggr <- motifdf %>% dplyr::select(!c(unselect_col,"name","baseMean","log2FoldChange","lfcSE","stat","pvalue","padj","gene_biotype","mgi_symbol")) %>% { if(weightby!=FALSE) mutate(., across(where(is.numeric), ~ (.x * get(weightby)))) else .} %>% # weight features by score dplyr::select(!any_of(as.character(weightby))) %>% # then we can drop the score since it will be nonsensical after the aggregation anyways group_by(label,seqnames,anno) %>% dplyr::summarise(across( where(is.numeric), .fns=sum )) %>% # sum up genewise feature counts ungroup() } } else { # loop for prox based assignments motifdf_aggr <- motifdf %>% dplyr::select(!c("name","baseMean","log2FoldChange","lfcSE","stat","pvalue","padj","gene_biotype")) %>% group_by(label,seqnames,anno) %>% dplyr::summarise(across( where(is.numeric), .fns=sum )) %>% # sum up genewise feature counts ungroup() } return(motifdf_aggr) } get_feature_and_label_metrics <- function(motifdf, trainvalidx, genenames){ targets_train <- motifdf[ trainvalidx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() targets_test <- motifdf[ -trainvalidx, ] %>% pull(label) %>% as.numeric(levels(.))[.] %>% as.matrix() metrics = data.frame( n_input_features = ncol( motifdf[ , -c(1,2,3)]),# first three columns are labels, chromosome and gene annotation n_neg_inst_train = table(targets_train)[['0']], n_pos_inst_train = table(targets_train)[['1']], n_neg_inst_test = table(targets_test)[['0']], n_pos_inst_test = table(targets_test)[['1']] ) return(metrics) } #------------------------------------------------ ## initialize object to gather all metrics #------------------------------------------------ all_metrics=data.frame( n_input_features=numeric(), n_neg_inst_train=numeric(), n_pos_inst_train=numeric(), n_neg_inst_test=numeric(), n_pos_inst_test=numeric() ) #------------------------------------------------ ## run proximity based #------------------------------------------------ #---------------------- # This is independent of the ABC results (no need to loop through different assignment variations) print("Getting metrics of prox-based model") my_rdsfile <- paste0(opt$featuredir,"prox.rds") if(!file.exists(my_rdsfile)){ motifdata_aggr_prox <- merge_motifdata_with_assignments(motifcounts_summitregion, assignment_summit_prox, contrast_DexLPSvLPS, maxonly="prox", excludepromoters=FALSE, weightby = FALSE, sepPromEnh = FALSE) saveRDS(motifdata_aggr_prox,file=my_rdsfile) } else{ motifdata_aggr_prox <- readRDS(my_rdsfile) } motifdata_aggr_prox_scaled <- motifdata_aggr_prox %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) motifdata_aggr_prox_tranval_idx <- motifdata_aggr_prox_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) metrics_prox <- get_feature_and_label_metrics (motifdata_aggr_prox_scaled, motifdata_aggr_prox_tranval_idx, genenames=genenames) all_metrics <- rbind(all_metrics, "prox"=metrics_prox) #---------------------- not_all_na <- function(x) any(!is.na(x)) for (motifdata in c("motifcounts_abcregion","motifcounts_summitregion")){ for(excludepromoters in c(FALSE,"all","onlyNONself")){ for (onlymax in c(TRUE,FALSE)){ for (sepPromEnh in c(TRUE,FALSE)){ for (weight in c(FALSE, "abcscore")){ # set assignments fitting for the input data if(motifdata=="motifcounts_abcregion"){ assignment_dexlps <- assignment_abcregion_dexlps assignment_lps <- assignment_abcregion_lps motifcounts_dexlps <- motifcounts_abcregion_dexlps motifcounts_lps <- motifcounts_abcregion_lps } else if (motifdata=="motifcounts_summitregion" ){ # in this case the motifcounts for the 2 conditions are the same, but there assignments differ assignment_dexlps <- assignment_summits_abcregion_dexlps assignment_lps <- assignment_summits_abcregion_lps motifcounts_dexlps <- motifcounts_summitregion motifcounts_lps <- motifcounts_summitregion } else {break} #-----------------------DEXLPS------------------------------ modelname <- paste(motifdata,"condition_dexlps_exclprom",excludepromoters,"onlymax",onlymax,"sepPromEnh",sepPromEnh,"weight",weight, sep="_") print(modelname) my_rdsfile <- here( paste0(opt$featuredir, modelname,".rds")) if(!file.exists(my_rdsfile)){ featurematrix_dexlps <- merge_motifdata_with_assignments(motifcounts_dexlps, assignment_dexlps, contrast_DexLPSvLPS, weightby=weight, maxonly=onlymax, excludepromoters=excludepromoters, sepPromEnh=sepPromEnh) saveRDS(featurematrix_dexlps, my_rdsfile) } else{ featurematrix_dexlps <- readRDS(my_rdsfile) } # remove columns that are NA after scaling featurematrix_dexlps_scaled <- featurematrix_dexlps %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) %>% dplyr::select(where(not_all_na)) motifdata_aggr_tranval_idx <- featurematrix_dexlps_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) new_metric <- get_feature_and_label_metrics (featurematrix_dexlps_scaled, motifdata_aggr_tranval_idx, genenames=genenames) all_metrics <- rbind(all_metrics, new_metric)%>% magrittr::set_rownames(c(rownames(all_metrics),modelname)) #--------------------------LPS--------------------------- modelname <- paste(motifdata,"condition_lps_exclprom",excludepromoters,"onlymax",onlymax,"sepPromEnh",sepPromEnh,"weight",weight, sep="_") print(modelname) my_rdsfile <- here(paste0(opt$featuredir, modelname,".rds")) if(!file.exists(my_rdsfile)){ featurematrix_lps <- merge_motifdata_with_assignments(motifcounts_lps, assignment_lps, contrast_DexLPSvLPS, weightby=weight, maxonly=onlymax, excludepromoters=excludepromoters, sepPromEnh=sepPromEnh) saveRDS(featurematrix_lps,file=my_rdsfile) } else{ featurematrix_lps <- readRDS(my_rdsfile) } featurematrix_lps_scaled <- featurematrix_lps %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) %>% dplyr::select(where(not_all_na)) motifdata_aggr_tranval_idx <- featurematrix_lps_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) new_metric <- get_feature_and_label_metrics (featurematrix_lps_scaled, motifdata_aggr_tranval_idx, genenames=genenames) all_metrics <- rbind(all_metrics, new_metric)%>% magrittr::set_rownames(c(rownames(all_metrics),modelname)) #-----------------------DIFFERENCE------------------------------ modelname <- paste(motifdata,"condition_DexLPS-LPS_exclprom",excludepromoters,"onlymax",onlymax,"sepPromEnh",sepPromEnh,"weight",weight, sep="_") print(modelname) my_rdsfile <- here(paste0(opt$featuredir, modelname,".rds")) if(!file.exists(my_rdsfile)){ # they contain the same motifs, but not the same genes. # doublecheck that all columnnames are identical table(colnames(featurematrix_dexlps) == colnames(featurematrix_lps)) # motifcounts missing in one condition should be set to 0 merged_featurematrix <- merge(featurematrix_dexlps, featurematrix_lps, by=c("anno","label","seqnames"), all=TRUE) # replace missing counts in one of the conditions with 0 merged_featurematrix[is.na(merged_featurematrix)] <- 0 # use dataframe suffix to grab respective columns featurematrix_diff <- merged_featurematrix[grep(".x$",colnames(merged_featurematrix))] - merged_featurematrix[grep(".y$",colnames(merged_featurematrix))] # tidy up column names colnames(featurematrix_diff) <- gsub(".x$","", colnames(featurematrix_diff)) # add first 3 columns back after computing the difference of the counts featurematrix_diff <- cbind(merged_featurematrix[1:3],featurematrix_diff) saveRDS(featurematrix_diff,file=my_rdsfile) } else{ featurematrix_diff <- readRDS(my_rdsfile) } featurematrix_diff_scaled <- featurematrix_diff %>% mutate(., across(where(is.numeric), ~(scale(.) %>% as.vector))) %>% dplyr::select(where(not_all_na)) # run GLM and look at performance featurematrix_diff_tranval_idx <- featurematrix_diff_scaled %>% with(which(seqnames!="chr1" & seqnames!="chr8" & seqnames!="chr9")) new_metric <- get_feature_and_label_metrics (featurematrix_diff_scaled, featurematrix_diff_tranval_idx, genenames=genenames) all_metrics <- rbind(all_metrics, new_metric)%>% magrittr::set_rownames(c(rownames(all_metrics),modelname)) } } } } } all_metrics <- all_metrics %>% mutate(ratio_pos_inst_train = n_pos_inst_train/(n_pos_inst_train+n_neg_inst_train), ratio_pos_inst_test = n_pos_inst_test/(n_pos_inst_test+n_neg_inst_test) ) # let'S add the number of non-zero coefficients to this model_coefs_joint <- readRDS( opt$model_coefs_joint ) model_coefs_sep <- readRDS( opt$model_coefs_sep ) nonzero_columentries <- rbind ( colSums(model_coefs_joint != 0, na.rm=TRUE) %>% as.data.frame(), colSums(model_coefs_sep != 0, na.rm=TRUE) %>% as.data.frame() ) colnames(nonzero_columentries) <- "n_sel_features_inclintercept" all_metrics <- merge(all_metrics, nonzero_columentries, by="row.names") %>% relocate(n_sel_features_inclintercept, .after = n_input_features ) # parse feature engineering choices from modelname and add them as variables all_metrics <- all_metrics %>% mutate(condition = case_when(grepl("condition_lps", Row.names) ~ "LPS", grepl("condition_DexLPS-LPS", Row.names) ~ "Dex+LPS - LPS", grepl("condition_dexlps", Row.names) ~ "Dex+LPS"), input_region = case_when(grepl("motifcounts_summitregion", Row.names) ~ "GR summitregions", grepl("motifcounts_abcregion", Row.names) ~ "active regions"), excludepromoters = case_when(grepl("exclprom_all", Row.names) ~ "all", grepl("exclprom_onlyNONself", Row.names) ~ "nonself", grepl("exclprom_FALSE", Row.names) ~ "none"), mapping = case_when(grepl("onlymax_FALSE", Row.names) ~ "1-to-many (ABC-based)", grepl("onlymax_TRUE", Row.names) ~ "1-to-1 (ABC-based)"), weight = case_when(grepl("weight_abcscore", Row.names) ~ "abcscore", grepl("weight_FALSE", Row.names) ~ "none"), prom_and_enh_features = case_when(grepl("sepPromEnh_TRUE", Row.names) ~ "separate", grepl("sepPromEnh_FALSE", Row.names) ~ "aggregate") ) # manually update choices for the reference model all_metrics <- all_metrics %>% rows_update( tibble(Row.names = "prox", condition="Dex+LPS", input_region="GR summitregions", mapping="1-to-1 (proximity-based)", weight="none"), by = "Row.names") summary(all_metrics) summary(all_metrics$ratio_pos_inst_train-all_metrics$ratio_pos_inst_test) write.table(all_metrics, file=opt$outfile, quote=FALSE, sep="\t", row.names = FALSE, col.names = TRUE ) |
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 | set -eo pipefail # for the input samples indir_ctrl=$1 indir_samples=$2 outdir=$3 ln -s "$(realpath $indir_ctrl/GAR0517/fastq/GAR0517_BC7FEMANXX_AGTCAA_L007_R1_001.fastq.gz)" $outdir/input_rep1_R1.fastq.gz ln -s "$(realpath $indir_ctrl/GAR0517/fastq/GAR0517_BC7FEMANXX_AGTCAA_L007_R2_001.fastq.gz)" $outdir/input_rep1_R2.fastq.gz ln -s "$(realpath $indir_ctrl/Sample_MUC9117/MUC9117_R1_merged.fastq.gz)" $outdir/input_rep2_R1.fastq.gz ln -s "$(realpath $indir_ctrl/Sample_MUC9117/MUC9117_R2_merged.fastq.gz)" $outdir/input_rep2_R2.fastq.gz ln -s "$(realpath $indir_ctrl/Sample_MUC9118/MUC9118_R1_merged.fastq.gz)" $outdir/input_rep3_R1.fastq.gz ln -s "$(realpath $indir_ctrl/Sample_MUC9118/MUC9118_R2_merged.fastq.gz)" $outdir/input_rep3_R2.fastq.gz ln -s "$(realpath $indir_ctrl/Sample_GAR1531/GAR1531_S13_L002_R1_001.fastq.gz)" $outdir/input_rep4_R1.fastq.gz ln -s "$(realpath $indir_ctrl/Sample_GAR1531/GAR1531_S13_L002_R2_001.fastq.gz)" $outdir/input_rep4_R2.fastq.gz # for the GR 2020 GR samples ln -s "$(realpath $indir_samples/Sample_MUC20387/MUC20387_S6_R1_001.fastq.gz)" $outdir/DexLPS_chipseq_GR_rep1_R1.fastq.gz ln -s "$(realpath $indir_samples/Sample_MUC20387/MUC20387_S6_R2_001.fastq.gz)" $outdir/DexLPS_chipseq_GR_rep1_R2.fastq.gz ln -s "$(realpath $indir_samples/Sample_MUC20388/MUC20388_S7_R1_001.fastq.gz)" $outdir/DexLPS_chipseq_GR_rep2_R1.fastq.gz ln -s "$(realpath $indir_samples/Sample_MUC20388/MUC20388_S7_R2_001.fastq.gz)" $outdir/DexLPS_chipseq_GR_rep2_R2.fastq.gz |
2 3 4 5 6 7 8 9 10 11 12 13 14 15 | set -eo pipefail # for the input samples indir=$1 outdir=$2 ln -s "$(realpath $indir/GAR0814_S7_L002_R1_001.fastq.gz)" $outdir/LPS_histone_H3K27ac_rep1_R1.fastq.gz ln -s "$(realpath $indir/GAR0814_S7_L002_R2_001.fastq.gz)" $outdir/LPS_histone_H3K27ac_rep1_R2.fastq.gz ln -s "$(realpath $indir/GAR0815_S8_L002_R1_001.fastq.gz)" $outdir/LPS_histone_H3K27ac_rep2_R1.fastq.gz ln -s "$(realpath $indir/GAR0815_S8_L002_R2_001.fastq.gz)" $outdir/LPS_histone_H3K27ac_rep2_R2.fastq.gz ln -s "$(realpath $indir/GAR0816_S9_L002_R1_001.fastq.gz)" $outdir/DexLPS_histone_H3K27ac_rep1_R1.fastq.gz ln -s "$(realpath $indir/GAR0816_S9_L002_R2_001.fastq.gz)" $outdir/DexLPS_histone_H3K27ac_rep1_R2.fastq.gz ln -s "$(realpath $indir/GAR0823_S10_L002_R1_001.fastq.gz)" $outdir/DexLPS_histone_H3K27ac_rep2_R1.fastq.gz ln -s "$(realpath $indir/GAR0823_S10_L002_R2_001.fastq.gz)" $outdir/DexLPS_histone_H3K27ac_rep2_R2.fastq.gz |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 | suppressPackageStartupMessages(library(optparse, warn.conflicts=F, quietly=T)) option_list <- list( make_option(c("--gene_annot"), type="character", help="Path to gencode annotation used to retrieve promoter regions"), make_option(c("--binding_sites_remap"), type="character", help="Path to peaks from the 2022 mouse remap release filtered for macrophages"), make_option(c("--rna_nascent_fpkm"), type="character", help="Path to normalized expression counts of nascent samples"), make_option(c("--genekey"), type="character", help="Path to genekey used to map ensembl geneIDs to mgi symbols"), make_option(c("--cage"), type="character", help="Path to bed file with location of max score within each cage read cluster"), make_option(c("--outdir"), type="character", help="Path to output directory") ) opt <- parse_args(OptionParser(option_list=option_list)) dir.create( opt$outdir ) # created the first time executing ftfbs_tss_annot_rds <- paste0( opt$outdir, "tfbs_tss2000u200d_annot.rds") ftfbs_tssCAGE_annot_rds <- paste0( opt$outdir, "tfbs_tssCAGE2000u200d_annot.rds") suppressPackageStartupMessages(library(here, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(dplyr, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(ggplot2, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(plsgenomics, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(data.table, warn.conflicts=F, quietly=T)) suppressPackageStartupMessages(library(GenomicRanges, warn.conflicts=F, quietly=T)) # ------------------------------------------------------------------------------ print("Get TFBS to TSS assignments.") # ------------------------------------------------------------------------------ # MAC specific TSS mac_spec_tss <- rtracklayer::import.bed(opt$cage) mac_spec_tss <- mac_spec_tss %>% filter(!duplicated(mac_spec_tss$SYMBOL)) %>% plyranges::as_granges() load_gene_annotation <- function(gene_annot) { # load gene annotation ga <- fread(gene_annot) # file format is: chr origin type start stop U strand U add_info colnames(ga) <- c("chr", "origin", "type", "start", "stop", "score", "strand", "frame", "info") # extract ranges ra <- with(ga, GRanges(chr, IRanges(start, stop), strand)) # extract the additional attributes and merge with ranges object attrs <- strsplit(ga$info, ";") gene_id <- sapply(attrs, function(x) { sapply(strsplit(x[grepl("gene_id",x)], " "), "[[", 2) }) gene_name <- sapply(attrs, function(x) { sapply(strsplit(x[grepl("gene_name",x)], " "), "[[", 3) }) gene_biotype <- sapply(attrs, function(x) { sapply(strsplit(x[grepl("gene_type",x)], " "), "[[", 3) }) # remove any lingering quotes gene_id <- gsub("\"", "", gene_id) gene_name <- gsub("\"", "", gene_name) gene_biotype <- gsub("\"", "", gene_biotype) # add to ranges object names(ra) <- gene_id ra$SYMBOL <- gene_name ra$BIOTYPE <- gene_biotype # finally, filter out 'misc_RNA' types and unusual chromosomes ra <- ra[ra$BIOTYPE != "misc_RNA"] ra <- keepStandardChromosomes(ra) return(ra) } annotate_tfbs_to_tss <- function(binding_sites_remap, tss) { # get the TFBS regions from remap tfbs = rtracklayer::import(binding_sites_remap) ann = t(matrix(unlist(strsplit(values(tfbs)[,"name"], ",", fixed=T)), nrow=3)) colnames(ann) = c("geo_id", "TF", "condition") values(tfbs) = DataFrame(name=values(tfbs)[,"name"], data.frame(ann, stringsAsFactors=F)) # create an annotation matrix for the TSS chip = paste(values(tfbs)[,"TF"], values(tfbs)[,"condition"], sep=".") chip_exp = unique(chip) tfbs_ann = sapply(chip_exp, function(x) overlapsAny(tss, tfbs[chip == x])) rownames(tfbs_ann) = names(tss) return(tfbs_ann) } # TSS only from ref #------------------ ga <- load_gene_annotation(opt$gene_annot) tss <- promoters(ga, 2000, 200) names(tss) <- tss$SYMBOL tfbs_annot <- annotate_tfbs_to_tss(opt$binding_sites_remap, tss) saveRDS(tfbs_annot, file=ftfbs_tss_annot_rds) # TSS from CAGE and only from ref where we have none #------------------ ga_noCAGE <- ga %>% filter(!SYMBOL %in% mac_spec_tss$SYMBOL) # add info on those genes where we don't have mac specific TSS tss_wCAGE <- c(mac_spec_tss,ga_noCAGE) tss_wCAGE <- promoters(tss_wCAGE, 2000, 200) names(tss_wCAGE) <- tss_wCAGE$SYMBOL tfbs_annot_CAGE <- annotate_tfbs_to_tss(opt$binding_sites_remap, tss_wCAGE) saveRDS(tfbs_annot_CAGE, file=ftfbs_tssCAGE_annot_rds) # ------------------------------------------------------------------------------ print("Read in expression data.") # ------------------------------------------------------------------------------ expr_fpkm <- read.table(opt$rna_nascent_fpkm, header=TRUE) expr_fpkm_nototal <- expr_fpkm [ !grepl('total|Total', colnames(expr_fpkm))] # ------------------------------------------------------------------------------ print("Get ensembl to MGI mapping") # ------------------------------------------------------------------------------ geneKey <- read.delim(opt$genekey) # ------------------------------------------------------------------------------ print("Change gene annotation from Ensembl to MGI") # ------------------------------------------------------------------------------ expr <- merge(expr_fpkm_nototal, geneKey[,c("ensembl_gene_id", "mgi_symbol")], by.x='Geneid', by.y='ensembl_gene_id') expr <- expr %>% filter(mgi_symbol!="") # summarize the expression values for those mgi symbols that have multiple entries expr <- expr %>% group_by(mgi_symbol) %>% summarise(across(2:(ncol(expr)-1), mean)) %>% tibble::column_to_rownames("mgi_symbol") write.table(expr, file=paste0( opt$outdir,"FPKMcounts_mgiaggr.tsv"), row.names = TRUE, col.names = TRUE, sep = "\t", quote = FALSE) # ------------------------------------------------------------------------------ print("Define annotation and data subsets.") # ------------------------------------------------------------------------------ tfs <- unique(sapply(strsplit(colnames(tfbs_annot), "\\."), "[[", 1)) # one TF might have the same target measured more than once -> summarize tss_annot_summarized <- sapply(tfs, function(tf) { rowSums(tfbs_annot[,grepl(paste0(tf, "\\."), colnames(tfbs_annot)), drop=F]) }) # get TFs and their targets (which targets are expressed) targets <- intersect(rownames(tfbs_annot), rownames(expr)) # we skip the filtering step for TFS that are expressed, and work with all of them for now # (partially because the genenames and TF protein names might not even match and we'll look at the results in more detail afterwards anyways) #tf_sub <- tfs[tfs %in% rownames(expr)] # get the annotation and data subsets annot_sub <- tss_annot_summarized[targets,,drop=F] #annot_sub <- annot_sub[, tf_sub] data_sub <- expr[targets,] # ------------------------------------------------------------------------------ print("Estimating TFAs using PLS/SIMPLS and substituting.") # ------------------------------------------------------------------------------ TFA <- plsgenomics::TFA.estimate(annot_sub, data_sub)$TFA rownames(TFA) <- colnames(annot_sub) colnames(TFA) <- colnames(data_sub) # ------------------------------------------------------------------------------ print("Saving results.") # ------------------------------------------------------------------------------ write.table(file=paste0( opt$outdir, "TFA_4sU.tsv"), TFA, sep="\t", quote=FALSE, col.names = TRUE, row.names = TRUE) # ------------------------------------------------------------------------------ print("Same steps with CAGE specific data") # ------------------------------------------------------------------------------ # ------------------------------------------------------------------------------ print("Define annotation and data subsets.") # ------------------------------------------------------------------------------ tfs_CAGE <- unique(sapply(strsplit(colnames(tfbs_annot_CAGE), "\\."), "[[", 1)) # one TF might have the same target measured more than once -> summarize tss_CAGE_annot_summarized <- sapply(tfs_CAGE, function(tf) { rowSums(tfbs_annot_CAGE[,grepl(paste0(tf, "\\."), colnames(tfbs_annot_CAGE)), drop=F]) }) # get TFs and their targets (which targets are expressed) targets_CAGE <- intersect(rownames(tfbs_annot_CAGE), rownames(expr)) # we skip the filtering step for TFS that are expressed, and work with all of them for now # (partially because the genenames and TF protein names might not even match and we'll look at the results in more detail afterwards anyways) #tf_sub <- tfs[tfs %in% rownames(expr)] # get the annotation and data subsets annot_CAGE_sub <- tss_CAGE_annot_summarized[targets_CAGE,,drop=F] #annot_sub <- annot_sub[, tf_sub] data_sub_CAGE <- expr[targets_CAGE,] # ------------------------------------------------------------------------------ print("Estimating TFAs using PLS/SIMPLS and substituting.") # ------------------------------------------------------------------------------ TFA_CAGE <- plsgenomics::TFA.estimate(annot_CAGE_sub, data_sub_CAGE)$TFA rownames(TFA_CAGE) <- colnames(annot_CAGE_sub) colnames(TFA_CAGE) <- colnames(data_sub_CAGE) # ------------------------------------------------------------------------------ print("Saving results.") # ------------------------------------------------------------------------------ write.table(file=paste0( opt$outdir, "TFA_4sU_CAGE.tsv"), TFA_CAGE, sep="\t", quote=FALSE, col.names = TRUE, row.names = TRUE) |
R
ggplot2
dplyr
data.table
optparse
GenomicRanges
here
plsgenomics
From
line
1
of
scripts/tf_activity_4sU.r
21 22 23 24 25 26 27 28 | shell: """ mkdir -p {params.outdir} && \ wget {params.link1} -P {params.outdir} && \ zcat {params.temp_s1} | awk '{{print $1,$2,$6,$5}}' OFS='\t' - > {output.s1} && \ wget {params.link2} -P {params.outdir} && \ zcat {params.temp_s2} | awk '{{print $1,$2,$6,$5}}' OFS='\t' - > {output.s2} """ |
50 51 52 53 54 55 56 | shell: """ mkdir -p {params.outdir} && \ Rscript src/scripts/get_TSS_from_cage.r --ctss_pool1 {input.ctss_pool1} --ctss_pool2 {input.ctss_pool2} \ --liftoverchain {input.liftoverchain} --gencode_mm9_geneanno {input.gencode_mm9_geneanno} \ --gencode_mm10_geneanno {input.gencode_mm10_geneanno} --outdir {params.outdir} """ |
64 65 66 67 68 | shell: """ wget https://hicfiles.tc4ga.com/public/juicer/juicer_tools.1.9.9_jcuda.0.8.jar -P src/ && ln -s juicer_tools.1.9.9_jcuda.0.8.jar src/juicer_tools.jar """ |
80 81 82 83 84 85 86 87 88 | shell: """ mkdir -p {params.outdir} && \ python src/scripts/abcmodel/juicebox_dump.py \ --hic_file {input.hic} \ --juicebox "java -jar src/juicer_tools.jar" \ --outdir {params.outdir} \ --chromosomes 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,X """ |
100 101 102 103 104 105 106 107 108 109 | shell: """ python src/scripts/abcmodel/compute_powerlaw_fit_from_hic.py \ --hicDir {params.hicdir} \ --outDir {params.outdir} \ --maxWindow 1000000 \ --minWindow 5000 \ --resolution 5000 \ --chr $(echo chr{{1..19}} chrX | tr ' ' ,) """ |
119 120 121 122 123 | shell: """ mkdir -p results/current/atacseq/macs2/ && \ bedtools sort -faidx {input.chromsizes} -i {input.narrowpeak} > {output} """ |
132 133 | shell: "samtools merge {output} {input}/*.bam" |
142 143 | shell: "cat {input.mac_cage_tssclusterregions} {input.ref_promoterregions} {input.summitintervals_GR_ChIPseq} > {output}" |
161 162 163 164 165 166 167 168 169 170 171 | shell: """ python src/scripts/abcmodel/makeCandidateRegions.py \ --narrowPeak {input.narrowpeak_sorted} \ --bam {input.atac_bam} \ --outDir {params.outdir} \ --chrom_sizes {input.chromsizes} \ --regions_blocklist {input.blacklist} \ --regions_includelist {input.includeregions} \ --nStrongestPeaks 150000 """ |
188 189 190 191 192 | shell: """ mkdir -p "results/current/abcmodel/expression/" && \ Rscript src/scripts/get_TPM_condition_expression.r -g {input.genekey} -e {input.tpm} """ |
250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 | shell: """ python src/scripts/abcmodel/predict.py \ --enhancers {input.enhancerlist} \ --genes {input.genelist} \ --HiCdir {params.hic_dir} \ --chrom_sizes {input.chromsizes} \ --hic_resolution {params.hic_res} \ --scale_hic_using_powerlaw \ --threshold .02 \ --cellType {params.celltype} \ --outdir {params.outdir} \ --make_all_putative \ --chromosomes $(echo chr{{1..19}} chrX | tr ' ' ,) """ |
273 274 275 276 | shell: """ zcat {input} |awk '{{if ($21>=0.02 && $21!="NaN") {{print}}}}' FS='\\t' OFS='\t' > {output} """ |
291 292 293 294 295 296 297 | shell: """ mkdir -p $(dirname {output}) && \ perlbrew use perl-5.34.0 && \ Rscript src/scripts/memes_runanalyses_ABCenhancerregions.r \ --ABC_all {input.abc} --memedb_expressed {input.memedb} --output {output.rds_fimo} """ |
9 10 11 12 13 | shell: """ mkdir -p $(dirname {output}) && \ cat {input} > {output} """ |
38 39 40 41 42 | shell: """ mkdir -p {params.outdir} && \ sh src/scripts/symlink_GR_2020.sh {params.indir_ctrl} {params.indir_samples} {params.outdir} """ |
66 67 68 69 70 | shell: """ mkdir -p {params.outdir} && \ sh src/scripts/symlink_histone_samples.sh {params.indir} {params.outdir} """ |
87 88 89 90 91 | shell: """ mkdir -p {params.outdir} && \ sh src/scripts/symlink_test.sh {params.indir} {params.outdir} """ |
107 108 109 110 111 | shell: "mkdir -p results/current/ChIP/{wildcards.exp}/fastq/trimmed; \ cutadapt -m {params.minimum_fragment_length} \ -a AGATCGGAAGAGCACACGTCT \ -o {output} {input.fastq}" |
126 127 | shell: "multiqc results/current/logs results/current/ChIP/H3K27ac/QC --outdir results/current/ChIP/H3K27ac --force" |
147 148 149 150 151 152 153 | shell: "mkdir -p results/current/logs && \ mkdir -p results/current/ChIP/{wildcards.exp}/bam && \ gzip -cd {input[fastq]} | bowtie -S -p {params.nworkers} --chunkmbs 512 \ -k {params.alignments_reported} -m {params.multialignments_allowed} \ -v {params.mismatches_allowed} {params.index} - | \ samtools view -F 4 -Sbo {output} -" |
174 175 176 177 178 179 180 | shell: "mkdir -p results/current/logs && \ mkdir -p $(dirname {output}) && \ bowtie -S -p {params.nworkers} --chunkmbs 512 \ -k {params.alignments_reported} -m {params.multialignments_allowed} \ -v {params.mismatches_allowed} {params.index} \ -1 {input[fa1]} -2 {input[fa2]} | samtools view -F 4 -Sbo {output} -" |
196 197 198 199 200 201 202 203 204 | shell: """ mkdir -p {params.tempdir} && \ mkdir -p {params.outdir} && \ macs2 callpeak -t {input.sample} -c {input.control}\ --tempdir {params.tempdir} --qvalue 0.05 \ --keep-dup all --bdg --SPMR \ -g mm -f BAM --outdir {params.outdir} -n {params.name} """ |
215 216 217 218 219 220 221 222 223 | shell: """ mkdir -p {params.tempdir} && \ mkdir -p {params.outdir} && \ macs2 callpeak -t {input.sample} --broad \ --tempdir {params.tempdir} --qvalue 0.05 \ --keep-dup all --bdg --SPMR \ -g mm -f BAM --outdir {params.outdir} -n {params.name} """ |
241 242 243 244 245 246 | shell: "mkdir -p $(dirname {output[0]}) && \ mkdir -p {params.tempdir} && \ macs2 callpeak -t {input[0]} -c {input[1]} {input[2]} {input[3]} {input[4]}\ --tempdir {params.tempdir} --pvalue 0.1 \ --keep-dup 1 --bdg -g mm -f BAMPE --outdir {params.outdir} -n {params.name}" |
256 257 258 | shell: "mkdir -p $(dirname {output}) && \ idr --samples {input} --input-file-type narrowPeak --rank signal.value --idr-threshold 0.05 --output-file {output} --plot" |
269 270 | shell: "samtools merge {output} {input[0]} {input[1]}" |
281 282 283 284 285 286 287 288 | shell: """ cat {input.peaks1} {input.peaks2} | bedtools sort -i stdin | bedtools merge -i stdin > {output.peakuniverse} && \ echo {input.bam1} > {output.counts} && \ samtools view -c -L {output.peakuniverse} {input.bam1} >> {output.counts} && \ echo {input.bam2} >> {output.counts} && \ samtools view -c -L {output.peakuniverse} {input.bam2} >> {output.counts} """ |
299 300 301 302 303 304 305 306 | shell: """ cat {input.peaks1} {input.peaks2} | bedtools sort -i stdin | bedtools merge -i stdin > {output.peakuniverse} && \ echo {input.bam1} > {output.counts} && \ samtools view -c -L {output.peakuniverse} {input.bam1} >> {output.counts} && \ echo {input.bam2} >> {output.counts} && \ samtools view -c -L {output.peakuniverse} {input.bam2} >> {output.counts} """ |
15 16 17 18 19 | shell: """ Rscript src/scripts/DE_visualizations_4sU.r -n {input.norm} -a {input.metadata} --contrast {input.contrast} \ --log2fcthresh {params.log2fcthresh} -o {params.outdir} """ |
36 37 38 39 40 41 | shell: """ Rscript src/scripts/figure_chipseq_prepdata.r --chipseq_bam {input.bam} --chipseq_summits {input.summits} \ --nr3c1fullsitematches {input.nr3c1fullsitematches} --nr3c1halfsitematches {input.nr3c1halfsitematches} \ -o {params.outdir} """ |
66 67 68 69 70 | shell: """ Rscript src/scripts/figure_proxanno_prepdata.r --log2fcthresh {params.log2fcthresh} --chipseq_summits {input.summits} --genekey {input.genekey} \ --contrast_DexVSDexLPS {input.contrast} --meme_db_path {input.meme_db} --rna_nascent_fpkm {input.rna_nascent_fpkm} -o {params.outdir} """ |
85 86 87 88 89 90 | shell: """ Rscript src/scripts/figure_chipseq_plot.r --summitAnno {input.summitAnno} --chipseq_peaks {input.bed} --chipseq_summits {input.summits} \ --nr3c1fullsitematches {input.nr3c1fullsitematches} --nr3c1halfsitematches {input.nr3c1halfsitematches} \ --sm_summitranges {input.genomation_scorematrix} --streme {input.streme} """ |
110 111 112 113 114 115 | shell: """ Rscript src/scripts/figure_proxanno_plot.r --summitAnno_expr {input.summitAnno_expr} --summitAnno_df_expr {input.summitAnno_df_expr} \ --permtest_res {input.permtest_res} --fimo_results {input.fimo_results} --chipseq_summit_granges {input.chipseq_summit_granges} \ --deeptools {input.deeptools} --streme_100bp_up {input.streme_100bp_up} --streme_100bp_down {input.streme_100bp_down} """ |
128 129 130 131 132 | shell: """ Rscript src/scripts/abc_visualizations.r --ABC_DexLPS_all {input.ABC_DexLPS_all} --ABC_LPS_all {input.ABC_LPS_all} \ --contrast_DexVSDexLPS {input.contrast_DexVSDexLPS} --chipseq_summits {input.chipseq_summits} --igv {input.igv} """ |
157 158 159 160 161 162 | shell: """ Rscript src/scripts/abc_predictions_prepdata.r --ABC_DexLPS_all {input.ABC_DexLPS_all} --ABC_LPS_all {input.ABC_LPS_all} \ --fimo_results_dexlps {input.fimo_results_dexlps} --fimo_results_lps {input.fimo_results_lps} \ --fimo_results_summitregion {input.fimo_results_summitregion} --chipseq_ranges {input.chipseq_ranges} --outdir {params.outdir} """ |
186 187 188 189 190 191 192 | shell: """ Rscript src/scripts/abc_predictions.r --contrast_DexLPSvLPS {input.contrast_DexLPSvLPS} \ --assignment_summit_prox {input.assignment_summit_prox} --assignment_summits_abcregion_dexlps {input.assignment_summits_abcregion_dexlps} --assignment_summits_abcregion_lps {input.assignment_summits_abcregion_lps} \ --assignment_abcregion_dexlps {input.assignment_abcregion_dexlps} --assignment_abcregion_lps {input.assignment_abcregion_lps} \ --motifcounts_summitregion {input.motifcounts_summitregion} --motifcounts_abcregion_dexlps {input.motifcounts_abcregion_dexlps} --motifcounts_abcregion_lps {input.motifcounts_abcregion_lps} --outdir {params.outdir} """ |
207 208 209 210 211 212 | shell: """ Rscript src/scripts/figure_GLMs.r --model_coefs_joint {input.model_coefs_joint} \ --model_coefs_sep {input.model_coefs_sep} --auc {input.auc} \ --motifcounts_summitregion {input.motifcounts_summitregion} --raw_counts {input.raw_counts} """ |
225 226 227 228 229 230 | shell: """ Rscript src/scripts/figure_supplemental_GLMs.r --auc {input.auc_metrics} \ --dirname_featurematrizes {params.dirname_featurematrizes} \ --dirname_models {params.dirname_models} --outfig {output.png} """ |
247 248 249 250 251 252 | shell: """ Rscript src/scripts/tf_activity_4sU.r --gene_annot {input.gene_annot} \ --binding_sites_remap {input.binding_sites_remap} --rna_nascent_fpkm {input.rna_nascent_fpkm} \ --genekey {input.genekey} --cage {input.cage} --outdir {params.outdir} """ |
268 269 270 271 272 | shell: """ Rscript src/scripts/figure_stats.r --tfactivity {input.tfactivity} --expr {input.expr} --difffootprint {input.difffootprint} \ --memedb_expressed {input.memedb_expressed} --heatmap {input.heatmap} --chipms {input.chipms} """ |
296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 | shell: """ Rscript src/scripts/quantify_features_and_label_distribution.r --contrast_DexLPSvLPS {input.contrast_DexLPSvLPS} \ --assignment_summit_prox {input.assignment_summit_prox} \ --assignment_summits_abcregion_dexlps {input.assignment_summits_abcregion_dexlps} \ --assignment_summits_abcregion_lps {input.assignment_summits_abcregion_lps} \ --assignment_abcregion_dexlps {input.assignment_abcregion_dexlps} \ --assignment_abcregion_lps {input.assignment_abcregion_lps} \ --motifcounts_summitregion {input.motifcounts_summitregion} \ --motifcounts_abcregion_dexlps {input.motifcounts_abcregion_dexlps} \ --motifcounts_abcregion_lps {input.motifcounts_abcregion_lps} \ --model_coefs_joint {input.model_coefs_joint} \ --model_coefs_sep {input.model_coefs_sep} \ --featuredir {params.featuredir} --outfile {output} """ |
37 38 39 40 41 | shell: """ mkdir -p {params.outdir} && \ Rscript src/scripts/DE_analysis_4sU.r -c {input.counts} -a {input.metadata} -k 100 -o {params.outdir} """ |
61 62 63 64 65 66 | shell: """ Rscript src/scripts/custom_bw_libnorm.r \ --bw_DexLPS_h3k27ac {input.bw_DexLPS_h3k27ac} --bw_LPS_h3k27ac {input.bw_LPS_h3k27ac} --counts_h3k27ac {input.counts_h3k27ac} \ --bw_DexLPS_atac {input.bw_DexLPS_atac} --bw_LPS_atac {input.bw_LPS_atac} --counts_atac {input.counts_atac} --gtf {input.gtf} """ |
76 77 78 79 | shell: """ bigwigCompare --bigwig1 {input.bw1} --bigwig2 {input.bw2} -p 4 -o {output} """ |
89 90 91 92 | shell: """ bigwigCompare --bigwig1 {input.bw1} --bigwig2 {input.bw2} -p 4 -o {output} """ |
115 116 117 118 119 120 121 122 | shell: """ mkdir -p results/current/memes_bioc/ && \ perlbrew use perl-5.34.0 && \ Rscript src/scripts/memes_runanalyses.r \ --summit_granges {input.summit_granges} --memedb_expressed {input.memedb_expressed} && \ touch {output.done} """ |
135 136 137 138 139 140 141 | shell: """ mkdir -p $(dirname {output}) && \ perlbrew use perl-5.34.0 && \ Rscript src/scripts/memes_fimo_runanalyses.r \ --summit_granges {input.summit_granges} --memedb_expressed {input.memedb_expressed} """ |
154 155 156 157 158 159 | shell: """ mkdir -p "results/current/integration/deeptools" && computeMatrix reference-point --referencePoint center -b 1000 -a 1000 -bs 50 -R {input.tUP} {input.tDOWN} \ -S {input.chipseq} {input.atac_diff} {input.h3k27ac_diff} -o {output} """ |
168 169 170 171 172 173 174 175 176 177 | shell: """ plotHeatmap -m {input} \ --sortRegions descend --refPointLabel summit \ --samplesLabel chipseq_DexLPS atac_diff h3k27ac_diff \ --colorMap Greens RdYlBu_r RdYlBu_r \ --zMin 0 -2 -2 --zMax 0.8 1.8 1.8 \ --regionsLabel "regions of upregulated genes" "regions of downregulated genes" \ -out {output} """ |
186 187 188 189 190 191 192 193 194 195 | shell: """ plotHeatmap -m {input} \ --sortRegions descend --refPointLabel summit \ --samplesLabel chipseq_DexLPS atac_diff h3k27ac_diff \ --colorMap Greens RdYlBu_r RdYlBu_r \ --zMin 0 -2 -2 --zMax 0.8 1.8 1.8 \ --regionsLabel "regions of upregulated genes" "regions of downregulated genes" \ -out {output} """ |
205 206 207 208 209 | shell: """ Rscript src/scripts/filter_fimo_for_motif.r --fimo {input.fimo} --summit_granges {input.summit_granges} \ --motif_altname {wildcards.motif_altname} """ |
224 225 226 227 228 229 | shell: """ mkdir -p $(dirname {output}) && \ computeMatrix reference-point --referencePoint center -b 500 -a 500 -bs 2 -R {input.targets} \ -S {input.chipseq} {input.atac_dexlps} {input.atac_lps} {input.h3k27ac_dexlps} {input.h3k27ac_lps} -o {output} """ |
238 239 240 241 242 243 244 245 246 | shell: """ plotHeatmap -m {input} \ --sortRegions descend --refPointLabel summit \ --samplesLabel chipseq_DexLPS atac_dexlps atac_lps h3k27ac_dexlps h3k27ac_lps \ --colorMap Greens RdYlBu_r RdYlBu_r RdYlBu_r RdYlBu_r \ --zMin 0 0 0 0 0 --zMax 0.8 10 10 10 10 \ -out {output} """ |
9 10 11 | shell: "mkdir -p {wildcards.path1}/QC; \ fastqc -o {wildcards.path1}/QC -f fastq {input}" |
24 25 26 27 | shell: "mkdir -p results/current/tmp && \ samtools sort {input.bam} -T results/current/tmp/ --threads {threads} | \ bedtools intersect -v -abam stdin -b {input.blacklist} > {output}" |
35 36 37 | shell: "mkdir -p {wildcards.path1}/stats && \ picard MarkDuplicates INPUT={input} OUTPUT={output.bam} REMOVE_DUPLICATES=true METRICS_FILE={output.metric} VALIDATION_STRINGENCY=LENIENT PROGRAM_RECORD_ID='null'" |
47 48 49 | shell: "bamCoverage -b {input.bam} --normalizeUsing None \ --binSize 1 -p {params.processors} --outFileName {output[0]}" |
59 60 61 | shell: "bamCoverage -b {input.bam} --normalizeUsing None --filterRNAstrand forward \ --binSize 1 -p {params.processors} --outFileName {output[0]}" |
71 72 73 | shell: "bamCoverage -b {input.bam} --normalizeUsing None --filterRNAstrand reverse \ --binSize 1 -p {params.processors} --outFileName {output[0]}" |
96 97 98 99 100 101 102 103 | shell: "bamCoverage -b {input.bam} --normalizeUsing CPM \ --binSize 1 -p {params.processors} --outFileName {output[0]}" # normalization: CPM: Counts Per Million mapped reads # generate bed files from IDR peaks for motif analysis rule generate_bed_from_idr: input: "{path1}/idr/{path2}_idr0-05" |
107 108 | shell: "cut -f 1,2,3,4,5 {input} > {output}" |
117 118 | shell: "sort -k8,8nr {input} > {output}" |
126 127 | shell: "awk '{{print $1,$2+$10,$2+$10+1,$4,$5}}' FS='\\t' OFS='\t' {input} > {output}" |
137 138 139 140 | shell: "mkdir -p results/current/genomesize/; \ samtools faidx {input[fa]} && \ cut -f1,2 {input[fa]}.fai > {output[chromsizes]}" |
147 148 | shell: "awk '{{print $1,0,$2}}' FS='\\t' OFS='\t' {input} > {output}" |
158 159 160 161 162 | shell: """ bedtools slop -i {input[summit]} -g {input[chromsizes]} -b {wildcards.slop} | \ sort -k1,1 -k2,2n - > {output[bed]} """ |
171 172 173 174 | shell: """ bedtools merge -i {input.bed} -c 4 -o collapse -delim "|" > {output.bed} """ |
183 184 | shell: "samtools index {input} && mv {input}.bai {output}" |
Support
Do you know this workflow well? If so, you can
request seller status , and start supporting this workflow.
Created: 1yr ago
Updated: 1yr ago
Maitainers:
public
URL:
https://github.com/heiniglab/GR-mediated_gene_repression
Name:
gr-mediated_gene_repression
Version:
1
Downloaded:
0
Copyright:
Public Domain
License:
GNU General Public License v3.0
Keywords:
BSgenome.Mmusculus.UCSC.mm10
BSgenome.Mmusculus.UCSC.mm9
org.Mm.eg.db
TxDb.Mmusculus.UCSC.mm10.knownGene
macs2
biomaRt
universalmotif
BEDTools
Bowtie 2
CAGEr
ChIPpeakAnno
ChIPseeker
ComplexHeatmap
Consensus
Cutadapt
DeepTools
DESeq2
FastQC
FiMO
genomation
GenomicRanges
MEME
memes
MultiQC
Picard
plyranges
rtracklayer
SAMtools
Snakemake
topGO
circlize
data.table
dplyr
ggplot2
ggpubr
glmnet
here
optparse
plsgenomics
plyr
png
RColorBrewer
stringr
- Future updates
Related Workflows

ENCODE pipeline for histone marks developed for the psychENCODE project
psychip pipeline is an improved version of the ENCODE pipeline for histone marks developed for the psychENCODE project.
The o...

Near-real time tracking of SARS-CoV-2 in Connecticut
Repository containing scripts to perform near-real time tracking of SARS-CoV-2 in Connecticut using genomic data. This pipeli...

snakemake workflow to run cellranger on a given bucket using gke.
A Snakemake workflow for running cellranger on a given bucket using Google Kubernetes Engine. The usage of this workflow ...

ATLAS - Three commands to start analyzing your metagenome data
Metagenome-atlas is a easy-to-use metagenomic pipeline based on snakemake. It handles all steps from QC, Assembly, Binning, t...
raw sequence reads
Genome assembly
Annotation track
checkm2
gunc
prodigal
snakemake-wrapper-utils
MEGAHIT
Atlas
BBMap
Biopython
BioRuby
Bwa-mem2
cd-hit
CheckM
DAS
Diamond
eggNOG-mapper v2
MetaBAT 2
Minimap2
MMseqs
MultiQC
Pandas
Picard
pyfastx
SAMtools
SemiBin
Snakemake
SPAdes
SqueezeMeta
TADpole
VAMB
CONCOCT
ete3
gtdbtk
h5py
networkx
numpy
plotly
psutil
utils
metagenomics

RNA-seq workflow using STAR and DESeq2
This workflow performs a differential gene expression analysis with STAR and Deseq2. The usage of this workflow is described ...

This Snakemake pipeline implements the GATK best-practices workflow
This Snakemake pipeline implements the GATK best-practices workflow for calling small germline variants. The usage of thi...