-
Notifications
You must be signed in to change notification settings - Fork 2
/
funcs.R
664 lines (597 loc) · 23.7 KB
/
funcs.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
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
# Copyright 2015 Angela Yen
# This file is part of ChromDiff.
# ChromDiff is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
# ChromDiff is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with ChromDiff. If not, see <http://www.gnu.org/licenses/>.
source("setvars.R", chdir=T)
suppressMessages(library(methods, quietly=TRUE))
suppressMessages(library(utils, quietly=TRUE))
## if no data, return NA
my.which.max <- function(x) {
if(length(which.max(x))==0) {
return(NA)
} else {
return(which.max(x))
}
}
my.which.min <- function(x) {
if(length(which.min(x))==0) {
return(NA)
} else {
return(which.min(x))
}
}
#get_corrected_logexp=function(group1, group2, genes) {
# orig.data=get_logexp(c(group1, group2), genes)
# group1.mean=mean(orig.data[group1, genes], na.rm=TRUE)
# group2.mean=mean(orig.data[group2, genes], na.rm=TRUE)
# final.mat=rbind(orig.data[group1, genes]-group1.mean, orig.data[group2, genes] - group2.mean)
# return(final.mat)
#}
get_random_logexp=function(celltype.order, numgenes) {
logexp.mat=get_all_logexp()
random.cols=sample(1:ncol(logexp.mat), numgenes)
random.genes=colnames(logexp.mat)[random.cols]
return(get_logexp(celltype.order, random.genes))
}
get_other_genes=function(genes.to.skip) {
logexp.mat=get_all_logexp()
allgenes=colnames(logexp.mat)
genes.to.keep=allgenes[which(!(allgenes %in% genes.to.skip))]
return(genes.to.keep)
}
get_other_logexp=function(celltype.order, genes.to.skip) {
genes.to.keep=get_other_genes(genes.to.skip)
return(get_logexp(celltype.order, genes.to.keep))
}
## get log expression for requested ensembl genes and celltypes
get_logexp=function(celltype.order, ensembl.geneorder) {
logexp.mat=get_all_logexp()
logexp.mat=fill_nas(logexp.mat, celltype.order, ensembl.geneorder)
return(logexp.mat)
}
fill_nas = function(mat, out.rownames, out.colnames) {
default=NA
## fill in colnames
na.colnames=out.colnames[which(!out.colnames %in% colnames(mat))]
fill.cols=matrix(rep(default, length(na.colnames)*nrow(mat)), nrow=nrow(mat), dimnames=list(rownames(mat), na.colnames))
mat=cbind(mat, fill.cols)
na.rownames=out.rownames[which(!out.rownames %in% rownames(mat))]
fill.rows=matrix(rep(default, length(na.rownames)*ncol(mat)), ncol=ncol(mat), dimnames=list(na.rownames, colnames(mat)))
mat=rbind(mat, fill.rows)
result.mat=mat[out.rownames, out.colnames]
return(result.mat)
}
## given a matrix, generate the dendrogram for it for dimension "dim", save it in savefile
load_or_get_dend <- function(savefile, mat, dim, narm, dend.str) {
if(any(dim(mat)==0)) { print("Warning: matrix passed in has dimension of 0"); return();}
if(dim(mat)[dim]==1) { stop(paste("Can not get dend for matrix with length of 1 for dimension", dim)) }
## if can load everything from saved file, do so and return
success=FALSE
if(file.exists(savefile)) {
print(paste("Loading", savefile, "..."))
load(savefile)
if(exists(dend.str)) {
success=TRUE
}
## if file exists but has the wrong data stored, delete the file
if(!success) {
file.remove(savefile)
}
}
# if could not load information, re-calculate now
if(!success) {
#save(list=ls(all=TRUE), file="tmp.rdata")
mat=as.matrix(mat)
print("Recalculating dend...")
dend=get_dend(mat, dim, narm)
assign(dend.str, dend)
print("Saving dend info...")
parentdir=dirname(savefile)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
save(list=c(dend.str), file=savefile)
}
return(get(dend.str))
}
comp.dens.plot = function(savefile, group1.vals, group2.vals, group1, group2) {
suppressMessages(library("ggplot2"))
allvals=c(group1.vals, group2.vals)
labels=c(rep(group1, length(group1.vals)), rep(group2, length(group2.vals)))
df=data.frame(value=allvals, label=labels)
p=ggplot(df) + geom_density(aes(x = value, group=label, colour = label)) + ggtitle("Densities from a kernel density estimator")
ggsave(savefile)
}
rm.nas = function(vec) {
return(vec[which(!is.na(vec))])
}
## get appropriate dend via hclust, ordered by row or column mean (for dim 1 or 2, respectively)
get_dend <- function(mat, dim, narm) {
if(dim==1) {
transformfunc<-function(x) {return(x)}
meansfunc=rowMeans
} else if (dim==2) {
transformfunc<-function(x) {return(t(x))}
meansfunc=colMeans
} else {
stop(paste("unrecognized dimension ", dim, sep=""))
}
success=tryCatch({
reorder(as.dendrogram(hclust(dist(transformfunc(mat)))), meansfunc(mat, na.rm=narm))
TRUE
}, error=function(e) {
FALSE
})
if(!success) {
## replace NAs with 0's
mat[which(is.na(mat))]=0
}
dend=reorder(as.dendrogram(hclust(dist(transformfunc(mat)))), meansfunc(mat, na.rm=narm))
return(dend)
}
get_order <- function(mat, dim, narm) {
return(order.dendrogram(get_dend(mat, dim, narm)))
}
gen_colors <- function(min, med, max) {
range=max-min
scalingfactor=100/range
nbw=(floor((med-min)*scalingfactor))
nwr=floor((max-med)*scalingfactor)
bwcols=colorRampPalette(c("blue", "white"))(nbw)
wrcols=colorRampPalette(c("white", "red"))(nwr)
return(append(bwcols, wrcols))
}
get_features_from_genes = function(genes, featurenames) {
corresponding_genes=get_genes_only(featurenames)
assoc.feats=names(corresponding_genes[which(corresponding_genes %in% genes)])
return(assoc.feats)
}
pick_recurring_gene_feats=function(considered.feats, cutoff) {
genenames_count=table(get_genes_only(names(considered.feats)))
ord_genenames_count=sort(genenames_count, decreasing=TRUE, method="shell")
runsum=0
genes_to_use=c()
for(ind in 1:length(ord_genenames_count)) {
if(runsum>cutoff) {
break
}
runsum=runsum+ord_genenames_count[ind]
genes_to_use=append(genes_to_use, names(ord_genenames_count)[ind])
}
## get features associated with those genes
feats_to_keep=get_features_from_genes(genes_to_use, names(considered.feats))[1:cutoff]
return(feats_to_keep)
}
## returns matrix with features as rows and cell types as columns
get_residuals <- function(model, metric, dependentvariable, no_covariate_correction) {
if(dependentvariable=="special") {
dependentvariable="age"
}else if (dependentvariable=="specialgi") {
dependentvariable="anatomy"
}
## go through each celltype, read in corrected values (calculate corrected values if they don't exist yet)
matrix.vec=c()
prefix=get_prefix(metric, model)
metricdir=get_metric_subdir(metric)
suffix=get_suffix(metric)
## check if dependent variable is one of the covariates we correct for
if(toupper(dependentvariable) %in% get_covariates_to_correct()) {
varsuffix=paste(".skip", dependentvariable, sep="")
} else {
varsuffix=".skipnone"
}
residmat.rdatafile=paste("rdata/", model, "/", metricdir, "residmat", varsuffix, suffix, ".rdata", sep="")
success=FALSE
if(file.exists(residmat.rdatafile) && file.info(residmat.rdatafile)$size>0) {
print(paste("Loading residuals from ", residmat.rdatafile, "...", sep=""))
tryCatch( {load(residmat.rdatafile);
if(nrow(resids)<ncol(resids)) {resids=t(resids)};
if(all(is.na(resids)==FALSE)) {success=TRUE}
},
error=function(e) {print("Error reading residual matrix file...")})
}
#print(colnames(resids))
if(!success) {
featnames=get_featnames(model, metric)
allcelltypes.rdatafile=paste("rdata/", model, "/", metricdir, "all" , suffix, ".rdata", sep="")
matLoaded=FALSE
if(file.exists(allcelltypes.rdatafile) && file.info(allcelltypes.rdatafile)$size>0) {
print(paste("Loading feature values for all celltypes from", allcelltypes.rdatafile, "..."))
tryCatch( {load(allcelltypes.rdatafile);
## make sure that necessary variables exists, and make sure there were no errors with the matrix (there should be no NA elements)
if(exists("mat") && (metric!="perc" || exists("totalcounts")) && (all(is.na(mat)==FALSE))) {matLoaded=TRUE}},
error=function(e){print("Error reading all epigenomes files...")})
}
if(!matLoaded) {
firstcounts=TRUE
celltypes.list=get_all_celltypes()
print("Loading celltype data from each file...")
validcelltypes=c()
for (ind in 1:length(celltypes.list)) {
vecLoaded=FALSE
sharedrdatafile=paste("rdata/", model, "/", metricdir, celltypes.list[ind], suffix, ".rdata", sep="")
datafile=paste(prefix, celltypes.list[ind], suffix, ".txt", sep="")
## load or read in data of feature values for each celltype
if (file.exists(sharedrdatafile) && file.info(sharedrdatafile)$size>0) {
print(paste0("Loading rdata file for ", celltypes.list[ind], "..."))
load(sharedrdatafile)
if(all(is.na(curr.vec)==FALSE)) {
vecLoaded=TRUE
}
}
if (!vecLoaded) {
if (file.exists(datafile) && file.info(datafile)$size>0) {
print(paste0("Reading data file for ", celltypes.list[ind], "..."))
curr.vec=scan(datafile, quiet=TRUE)
parentdir=dirname(sharedrdatafile)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
save(curr.vec, file=sharedrdatafile)
} else { ## feature values txt file did not exist for this epigenome
warning(paste0(datafile, " not found. Skipping ", celltypes.list[ind], "..."))
next
}
}
validcelltypes=append(validcelltypes, celltypes.list[ind])
if(metric=="perc") {
countsfile=paste(prefix, celltypes.list[ind], "_counts.txt", sep="")
newtotalcounts=scan(countsfile, quiet=TRUE)
if(firstcounts) {
totalcounts=newtotalcounts
firstcounts=FALSE
} else {
if(!all(totalcounts==newtotalcounts)) {
stop(paste("Total counts of", celltypes.list[ind], "does not agree with previous counts"))
}
}
}
if(length(curr.vec)!=length(featnames)) {
stop(paste("Wrong number of feature numbers for ", celltypes.list[ind], ": expected ", length(featnames), " but saw ", length(curr.vec), sep=""))
}
matrix.vec=append(matrix.vec, curr.vec)
}
if(length(matrix.vec)==0) {
print("No state calls found for any cell type...")
quit()
}
## matrix has features as rows and cell types as columns
mat=matrix(matrix.vec, ncol=length(validcelltypes), dimnames=list(featnames, validcelltypes), byrow=FALSE)
parentdir=dirname(allcelltypes.rdatafile)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
validcelltypes.rdatafile=get_validcelltypesfile(model, metric, dependentvariable)
save(validcelltypes, file=validcelltypes.rdatafile)
if(metric=="perc") {
save(mat, totalcounts, file=allcelltypes.rdatafile)
} else {
save(mat, file=allcelltypes.rdatafile)
}
}
if(no_covariate_correction) {
return(mat)
}
## covariate mat has variables as columns and cell types as rows
##correct for all covariates EXCEPT current dependent variable
fixed.cov.mat <- get_covariate_mat(dependentvariable)
## pick out valid celltypes (listed in matrix of data) only
fixed.cov.mat = fixed.cov.mat[colnames(mat), ]
## invert matrix for helper function
inputmat=t(mat)
if(metric=="perc") {
resids=get_residuals_from_mats(fixed.cov.mat, inputmat, totalcounts, method="log")
} else {
resids=get_residuals_from_mats(fixed.cov.mat, inputmat, method="lin")
}
## we need resids to have same dimensions as original mat, so invert again
resids=t(resids)
print("Saving residuals...")
parentdir=dirname(residmat.rdatafile)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
save(resids, file=residmat.rdatafile)
}
return(resids)
}
## expects the same number of rows in cov.mat and values.mat, as they are the data being corrected (epigenomes)
## cov columns factors/covariates
## values.mat columns the values (genes) being corrected
get_residuals_from_mats = function(cov.mat, values.mat, method=c("log", "lin"), totalcounts=c()) {
if(method=="log") {
if(length(totalcounts)!=ncol(values.mat)) {
stop("Counts needed for logistic regression: length of totalcounts must match number of rows in values.mat")
}
}
if(nrow(cov.mat)!=nrow(values.mat)) {
stop("cov.mat and values.mat inputted into get_residuals_from_mats must have matching number of rows (epigenomes)")
}
## covariate matrix has celltypes as rows and factors/covariates as columns
if(ncol(cov.mat)>=1) {
if(method=="log") {
formstr=" success_fail ~ cov.mat[,1] "
} else if (method=="lin") {
formstr=" currvals ~ cov.mat[,1] "
}
for(colnum in 2:ncol(cov.mat)) {
formstr=paste(formstr, " + cov.mat[,", colnum, "]", sep="")
}
print(formstr)
print("Calculating residuals...")
## if any feature has any NA's, make the residual NAs
resids=matrix(rep(0, times=nrow(values.mat)*ncol(values.mat)), ncol=ncol(values.mat), dimnames=list(rownames(values.mat), colnames(values.mat)))
## correct each gene/region one at a time (each column)
for(i in 1:ncol(values.mat)) {
currvals=values.mat[,i]
if(length(which(is.na(currvals)))>0) {
result=(rep(NA, length(currvals)))
} else {
if(method=="log") {
counts=rep(totalcounts[i], times=length(currvals))
success_fail=cbind(round(currvals*counts), round((1-currvals)*counts))
result=(resid(glm(as.formula(formstr), family="binomial"), type="deviance"))
} else if(method=="lin"){
result=(resid(lm(as.formula(formstr))))
} else {
stop('Unrecognized method for get_residuals_from_mats: should be lin or log')
}
}
resids[,i]=result
if((i%%10000)==0) {
print(paste("Processing feature ", i, "...", sep=""))
}
}
} else {
resids=values.mat
}
return(resids)
}
load.background.vals <- function(metric, featnames, model) {
bgdir=paste("backgrounds/", model, "/", sep="")
if(metric=="windows" || metric=="deltawindows") {
fn="window_background"
} else if(metric=="deltas" || metric=="perc"){
fn="perc_background"
} else {
stop(paste("Unrecognized metric: ", metric))
}
rdatafile=paste(bgdir, fn, ".rdata", sep="")
if(file.exists(rdatafile) && file.info(rdatafile)$size>0) {
load(rdatafile)
} else {
fullfile=paste(bgdir, fn, ".txt", sep="")
bg.vals <- scan(fullfile)
names(bg.vals)=featnames
parentdir=dirname(rdatafile)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
save(bg.vals, file=rdatafile)
}
return(bg.vals)
}
## http://adv-r.had.co.nz/memory.html#garbarge-collection
mem <- function() {
bit <- 8L * .Machine$sizeof.pointer
if (!(bit == 32L || bit == 64L)) {
stop("Unknown architecture", call. = FALSE)
}
node_size <- if (bit == 32L) 28L else 56L
usage <- gc()
sum(usage[, 1] * c(node_size, 8)) / (1024 ^ 2)
}
comp.groups <- function(property, group.a, group.b, a.label, b.label, metrics="perc", tests="wilcox", test.corrections="fdr", model="core", no_covariate_correction=FALSE) {
pair.label=paste(a.label, ".", b.label, sep="")
## read gene names
genenames_file=paste0("genes/", generegions_label, "/genenames.txt")
genenames=scan(genenames_file, what=character())
## feature information
check_input(metrics, tests, test.corrections)
## start doing requested test and metric combinations
for (test in tests) {
test.func=get_test_func(test)
## repeat analysis for both percentage features and delta (diff from background) features
for(metric in metrics) {
prefix=get_prefix(metric, model)
suffix=get_suffix(metric)
full.rdatadir=get_full_rdatadir(model, test, metric)
dir.create(full.rdatadir, recursive=TRUE, showWarnings=FALSE)
for(correction in test.corrections) {
pv.file=paste(full.rdatadir, pair.label, ".pvals.rdata", sep="")
tmp.pv.file=paste(full.rdatadir, "fdr.", pair.label, ".pvals.rdata", sep="")
corr.func=get_corr_func(correction)
### Get corrected residuals (instead of raw data)
## residuals have celltypes as columns and features as rows
print("Getting residuals...")
resid.mat=get_residuals(model, metric, property, no_covariate_correction)
## pick out celltypes you are interested in
valid_celltypes_a=group.a[which(group.a %in% colnames(resid.mat))]
valid_celltypes_b=group.b[which(group.b %in% colnames(resid.mat))]
matrix.a=resid.mat[,valid_celltypes_a]
matrix.b=resid.mat[,valid_celltypes_b]
if(dim(matrix.a)[1]!=dim(matrix.b)[1]) {
stop("ERROR: number of features (rows) in matrices should match")
}
featnames=get_featnames(model, metric)
pvals.success=FALSE
## read pval from rdata file pv.file (NOTE: doesn't involve correction, because they are raw p-values, but for legacy reasons, we try tmp.pv.file which uses the pv.file with "fdr" in the filename)
if(file.exists(tmp.pv.file) && file.info(tmp.pv.file)$size>0) {
print(paste("Loading pvals from", tmp.pv.file, "..."))
tryCatch( {load(tmp.pv.file);
if(length(p.vals)==length(featnames)) {
pvals.success=TRUE
}},
error=function(e) {print("Error reading pval rdata file...")})
}
if(!pvals.success && file.exists(pv.file) && file.info(pv.file)$size>0){
print(paste("Loading pvals from", pv.file, "..."))
tryCatch( {load(pv.file);
if(length(p.vals)==length(featnames)) {
pvals.success=TRUE
}},
error=function(e) {print("Error reading pval rdata file...")})
}
if(!pvals.success) {
print("Calculating pvals...")
## do test for each feature, remember the indices of features with significant values
p.vals=test.func(matrix.a, matrix.b)
parentdir=dirname(pv.file)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
save(p.vals,file=pv.file)
pvals.success=TRUE
}
if(length(names(p.vals))==0) {
names(p.vals)=featnames
parentdir=dirname(pv.file)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
save(p.vals,file=pv.file)
}
## multiple hypothesis correction
corrected.pvals=corr.func(p.vals)
allpval.file=get_pval_file(model, metric, test, correction, property, a.label, b.label)
print(paste("Writing all pvals to ", allpval.file, "...", sep=""))
parentdir=dirname(allpval.file)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
write.table(corrected.pvals, quote=FALSE, col.names=FALSE, file=allpval.file)
## use significant corrected p-vals< .05 OR top 2000 if more than 2000 are significant
pvalcutoff=.05
cutoff=10000
all.sig.feats=names(which(corrected.pvals<pvalcutoff))
if(length(which(corrected.pvals<pvalcutoff))>cutoff) {
considered.feats=corrected.pvals[which(corrected.pvals<pvalcutoff)]
sig.feats=pick_recurring_gene_feats(considered.feats, cutoff)
} else {
sig.feats=names(which(corrected.pvals<=pvalcutoff))
}
## save significant features
if(length(sig.feats)>0) {
print(paste(length(sig.feats), "significant features found. Now visualizing..."))
sigpval.file=get_sigpval_file(model, metric, test, correction, property, a.label, b.label)
print(paste("Writing pvals to ", sigpval.file, "...", sep=""))
parentdir=dirname(sigpval.file)
dir.create(parentdir, recursive=TRUE, showWarnings=FALSE)
write.table(corrected.pvals[sig.feats], quote=FALSE, col.names=FALSE, file=sigpval.file)
} else {
print("No significant features found.")
}
}
}
}
}
reorder_get_vert_ints_height=function(colnames, dend.to.reorder, heightcutoff) {
tmp.colnames=colnames[order.dendrogram(dend.to.reorder)]
return(get_vert_ints_height(tmp.colnames, dend.to.reorder, heightcutoff) )
}
## returns a list of clusters, with each showing first and last element of the cluster
## use height cutoff
get_vert_ints_height = function(colnames, dend, heightcutoff) {
if(attributes(dend)$members==1) {
vert.ints=list(c(1,1))
} else {
## cut dendrogram at appropriate height
robust=cut(dend, heightcutoff)
nclust=length(robust$lower)
vert.ints=list()
for(ind in 1:nclust) {
curr.dend=robust$lower[[ind]]
## match dendrogram members (labels) to existing order of columns in matrix (colnames)
indices=match(labels(curr.dend), colnames)
## if there wer any problems, print info and stop
if(is.na(indices[1]) || !all(sort(indices)==(min(indices):max(indices)))) {
print(indices)
print(curr.dend)
print(colnames)
save(list=ls(all=TRUE), file="error.vert.rdata")
stop("Problem with getting vert indices: dend and colnames are not consistent");
}
## indices should be all consecutive
if(!all(sort(indices)==(min(indices):max(indices)))) {stop("Current ordering of column names conflicts with dendrogram used.")}
vert.ints[[ind]]=c(min(indices), max(indices))
#curr.sum=curr.sum+curr.clust.size
#vert.ints=append(vert.ints, curr.sum+.5)
}
}
return(vert.ints)
}
#get_vert_ints = function(colnames, pvclust, cutoff) {
# print(pvclust)
# if(nrow(pvclust$edges)==1) {
# vert.ints=list(c(1,1))
# } else {
# robust=pvpick(pvclust, alpha=cutoff)
# nclust=length(robust$clusters)
# vert.ints=list()
## for(ind in 1:nclust) {
# curr.clust=robust$clusters[[ind]]
# indices=match(curr.clust, colnames)
# ## if there wer any problems, print info and stop
# if(is.na(indices[1]) || !all(sort(indices)==(min(indices):max(indices)))) {
# print(indices)
# print(curr.clust)
# print(colnames)
# save(list=ls(all=TRUE), file="error.vert.rdata")
# stop("Problem with getting vert indices: pvclust and colnames are not consistent");
# }
# ## indices should be all consecutive
# if(!all(sort(indices)==(min(indices):max(indices)))) {stop("Current ordering of column names conflicts with dendrogram used.")}
# vert.ints[[ind]]=c(min(indices), max(indices))
# #curr.sum=curr.sum+curr.clust.size
# #vert.ints=append(vert.ints, curr.sum+.5)
# }
# }
# return(vert.ints)
#}
## get index of last instance of each element in tobematched in tosearch
match_last <- function(tobematched, tosearch) {
return(unlist(sapply(tobematched, function(tomatch){return(max(which(tomatch==tosearch)))})))
}
order.list.by.first=function(mylist) {
first.elts=sapply(mylist, function(x) x[1])
reord=order(first.elts)
new.list=lapply(reord, function(ind){return(vert.ints.noreps[[ind]])})
}
get_genes_only <- function(featnames) {
#print(head(featnames))
#split.mat=get_split_mat(featnames)
#return(split.mat[1,])
list.elts=strsplit(as.character(featnames), "_")
vec.of.genes=sapply(list.elts, function(x) {return(x[1])})
names(vec.of.genes)=featnames
return(vec.of.genes)
}
get_chrstates_only <- function(featnames) {
#split.mat=get_split_mat(featnames)
#return(split.mat[2,])
list.elts=strsplit(as.character(featnames), "_")
vec.of.states=sapply(list.elts, function(x) {return(x[2])})
names(vec.of.states)=featnames
return(vec.of.states)
}
get_split_mat <- function(featnames) {
split.mat=sapply(featnames, function(x) {return(strsplit(x, "_")[[1]])})
return(split.mat)
}
get.ranks.with.ties.for.nas <- function(vec) {
index.rank=rank(vec, na.last=TRUE)
na.inds=which(is.na(vec))
last.rank=length(vec)
index.rank[na.inds]=last.rank
return(index.rank)
}
get.state.ecdf <- function(ranks, totalfeatures, rawnum=FALSE){
xcoords=ranks
if(rawnum) {
ycoords=0:length(ranks)
} else {
ycoords=(0:length(ranks))/totalfeatures
}
return(stepfun(xcoords, ycoords))
}
plot_vector_as_hm<-function(outfile, data, rowlabels, collabels) {
pdf(outfile)
##image plots transpose so we transpose first
par( mar = par( "mar" ) + c( 2, 4, 0, 0 ) )
image(t(data), xaxt= "n", yaxt= "n" )
axis( 1, at=seq(0,1,length.out=ncol( data ) ), labels=collabels, las= 2, cex.axis=.5)
axis( 2, at=seq(0,1,length.out=nrow( data ) ), labels=rowlabels, las= 2, cex.axis=.7 )
dev.off()
}