forked from reubenthomas/PFOCRInPathwayAnalyses
-
Notifications
You must be signed in to change notification settings - Fork 1
/
ROC_curve.R
597 lines (428 loc) · 20.3 KB
/
ROC_curve.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
##########################################################################################knockout benchmark
require(pROC)
setwd("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/")
pvalue_results_human_voom<-readRDS("pvalue_results_human_voom.rds")
library(dplyr)
library(magrittr)
rsea_results_human_voom_go=list()
## process result.txt files for counts
for(i in 1:ncol(pvalue_results_human_voom)){
rsea_results_human_voom_go[[i]]<-read.table(paste0("GSE/",colnames(pvalue_results_human_voom)[i],"/rSEA/GO/result.txt"),header=TRUE)
}
knockout_target_gse<-read.table("knockout_target_gse",header=FALSE)
database_lists2<-load("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/pfocr_miny_April2021.RData")
database_lists1<-load("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/databases.RData")#has wp, pfocr, go
database_lists3<-load("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/databases_pfocr_3sets.RData")
min_set_size <- 3
max_set_size <- 500
database_lists2<- unname(unlist(sapply(database_lists2, grep, pattern="_list$", value = T, perl = T)))
for (db in database_lists2) {
eval(call("<-", as.name(db), Filter(Negate(is.null), lapply(get(db), function(x){
if(length(x) < min_set_size | length(x) > max_set_size)
NULL
else
x
}))
))
}
database_lists1 <- unname(unlist(sapply(database_lists1, grep, pattern="_list$", value = T, perl = T)))
for (db in database_lists1) {
eval(call("<-", as.name(db), Filter(Negate(is.null), lapply(get(db), function(x){
if(length(x) < min_set_size | length(x) > max_set_size)
NULL
else
x
}))
))
}
database_lists3<- unname(unlist(sapply(database_lists3, grep, pattern="_list$", value = T, perl = T)))
for (db in database_lists3) {
eval(call("<-", as.name(db), Filter(Negate(is.null), lapply(get(db), function(x){
if(length(x) < min_set_size | length(x) > max_set_size)
NULL
else
x
}))
))
}
########GO
merged_go<-read.table("/Users/mingyoungshin/git/PFOCRInPathwayAnalyses/human_knockout_go.txt",header=TRUE)
merged_go<-read.table("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/human_knockout_go_April2021version.txt",header=TRUE)
targets=which(is.na(merged_go$GO)==FALSE)#get GSEs matched with knock genes
target_gses=merged_go[targets,2]
length(target_gses)
target_gses=target_gses[-which(target_gses%in%c("GSE75291","GSE78506","GSE141126","GSE76421"))]
length(target_gses)
target_index=match(target_gses,colnames(pvalue_results_human_voom))#retrieve their index
target_index_go=target_index
rsea_go_sig_gene_count=NULL
rsea_go_aucs=NULL
rsea_go_aucs_ci=NULL
rsea_go_50lower=NULL
rsea_go_TRUEs=NULL
for(i in unique(target_index)){
print(i)
rsea_go_sig_gene_count=c(rsea_go_sig_gene_count,sum(rsea_results_human_voom_go[[i]]$Com.adjP <0.05,na.rm=TRUE))
if( sum(is.na(rsea_results_human_voom_go[[i]])) ==sum(lengths(rsea_results_human_voom_go[[i]])) ){next}#pass null cases
gse_match=which(as.character(merged_go$GSE)==colnames(pvalue_results_human_voom)[i])
for(j in gse_match){
print(merged_go$GO[j])
if(is.na(merged_go$GO[j])){next}
true_sig=merged_go$GO[j]
true_sig=as.character(strsplit(as.character(true_sig),";")[[1]])
true_sig=true_sig[true_sig%in%(names(go_list))]
print(length(true_sig))
response=apply(as.matrix(1:length(rsea_results_human_voom_go[[i]]$set_id)),1,function(x) rsea_results_human_voom_go[[i]]$set_id[x]%in%true_sig)
print(sum(response))
print(table(response))
rsea_go_TRUEs=rbind(rsea_go_TRUEs,c(i,merged_go$Symbol[j],as.numeric(table(response)['TRUE'])))
if(length(table(response))==1) next
roc_r=roc(response,as.numeric(rsea_results_human_voom_go[[i]]$Comp.adjP))
auc_r=auc(roc_r)
if(auc_r<=0.5){rsea_go_50lower=rbind(rsea_go_50lower,c(i,j))}
auc_r_ci=ci(auc_r)
if(!is.na(auc_r_ci[1])){
if(auc_r_ci[1] < roc_r$auc[1] & auc_r_ci[2] > roc_r$auc[1]){
rsea_go_aucs_ci=c(rsea_go_aucs_ci,auc_r[1])
}
}
rsea_go_aucs=c(rsea_go_aucs,auc_r[1])
}
}
boxplot(rsea_go_aucs)
########wp
rsea_results_human_voom_wp=list()
## process result.txt files for counts
for(i in 1:ncol(pvalue_results_human_voom)){
rsea_results_human_voom_wp[[i]]<-read.table(paste0("GSE/",colnames(pvalue_results_human_voom)[i],"/rSEA/WP/result.txt"),header=TRUE)
}
merged_wp<-read.table("/Users/mingyoungshin/git/PFOCRInPathwayAnalyses/human_knockout_wp.txt",header=TRUE)
merged_wp<-read.table("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/human_knockout_wp_April2021version.txt",header=TRUE)
targets=which(is.na(merged_wp$WP)==FALSE)#get GSEs matched with knock genes
target_gses=merged_wp[targets,2]
length(target_gses)
target_gses=target_gses[-which(target_gses%in%c("GSE75291","GSE78506","GSE141126","GSE76421"))]
length(target_gses)
target_index=match(target_gses,colnames(pvalue_results_human_voom))#retrieve their index
target_index_wp=target_index
rsea_wp_sig_gene_count=NULL
rsea_wp_aucs=NULL
rsea_wp_aucs_ci=NULL
rsea_wp_50lower=NULL
rsea_wp_TRUEs=NULL
for(i in unique(target_index)){
print(i)
rsea_wp_sig_gene_count=c(rsea_wp_sig_gene_count,sum(rsea_results_human_voom_wp[[i]]$Com.adjP <0.05,na.rm=TRUE))
if( sum(is.na(rsea_results_human_voom_wp[[i]])) ==sum(lengths(rsea_results_human_voom_wp[[i]])) ){next}#pass null cases
gse_match=which(as.character(merged_wp$GSE)==colnames(pvalue_results_human_voom)[i])
for(j in gse_match){
print(merged_wp$WP[j])
if(is.na(merged_wp$WP[j])){next}
true_sig=merged_wp$WP[j]
true_sig=as.character(strsplit(as.character(true_sig),";")[[1]])
true_sig=true_sig[true_sig%in%(names(wp_list))]
print(length(true_sig))
response=apply(as.matrix(1:length(rsea_results_human_voom_wp[[i]]$set_id)),1,function(x) rsea_results_human_voom_wp[[i]]$set_id[x]%in%true_sig)
print(sum(response))
print(table(response))
rsea_wp_TRUEs=rbind(rsea_wp_TRUEs,c(i,merged_wp$Symbol[j],as.numeric(table(response)['TRUE'])))
if(length(table(response))==1) next
roc_r=roc(response,as.numeric(rsea_results_human_voom_wp[[i]]$Comp.adjP))
auc_r=auc(roc_r)
if(auc_r<=0.5){rsea_wp_50lower=rbind(rsea_wp_50lower,c(i,j))}
auc_r_ci=ci(auc_r)
if(!is.na(auc_r_ci[1])){ if(auc_r_ci[1] < roc_r$auc[1] & auc_r_ci[2] > roc_r$auc[1]){
rsea_wp_aucs_ci=c(rsea_wp_aucs_ci,auc_r[1])
}}
rsea_wp_aucs=c(rsea_wp_aucs,auc_r[1])
}
}
boxplot(rsea_wp_aucs)
########pfocr
rsea_results_human_voom_pfocr=list()
## process result.txt files for counts
for(i in 1:ncol(pvalue_results_human_voom)){
rsea_results_human_voom_pfocr[[i]]<-read.table(paste0("GSE/",colnames(pvalue_results_human_voom)[i],"/rSEA/PFOCR/result.txt"),header=TRUE)
}
merged_pfocr<-read.table("/Users/mingyoungshin/git/PFOCRInPathwayAnalyses/human_knockout_pfocr.txt",header=TRUE)
merged_pfocr<-read.table("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/human_knockout_pfocr_April2021version.txt",header=TRUE)
targets=which(is.na(merged_pfocr$PFOCR)==FALSE)#get GSEs matched with knock genes
target_gses=merged_pfocr[targets,2]
length(target_gses)
target_gses=target_gses[-which(target_gses%in%c("GSE75291","GSE78506","GSE141126","GSE76421"))]
length(target_gses)
target_index=match(target_gses,colnames(pvalue_results_human_voom))#retrieve their index
target_index_pfocr=target_index
rsea_pfocr_sig_gene_count=NULL
rsea_pfocr_aucs=NULL
rsea_pfocr_aucs_ci=NULL
rsea_pfocr_50lower=NULL
rsea_pfocr_TRUEs=NULL
for(i in unique(target_index)){
print(i)
rsea_pfocr_sig_gene_count=c(rsea_pfocr_sig_gene_count,sum(rsea_results_human_voom_pfocr[[i]]$Com.adjP <0.05,na.rm=TRUE))
if( sum(is.na(rsea_results_human_voom_pfocr[[i]])) ==sum(lengths(rsea_results_human_voom_pfocr[[i]])) ){next}#pass null cases
gse_match=which(as.character(merged_pfocr$GSE)==colnames(pvalue_results_human_voom)[i])
for(j in gse_match){
print(merged_pfocr$PFOCR[j])
if(is.na(merged_pfocr$PFOCR[j])){next}
true_sig=merged_pfocr$PFOCR[j]
true_sig=as.character(strsplit(as.character(true_sig),";")[[1]])
true_sig=true_sig[true_sig%in%(names(pfocr_list))]
print(length(true_sig))
response=apply(as.matrix(1:length(rsea_results_human_voom_pfocr[[i]]$set_id)),1,function(x) rsea_results_human_voom_pfocr[[i]]$set_id[x]%in%true_sig)
print(sum(response))
print(table(response))
rsea_pfocr_TRUEs=rbind(rsea_pfocr_TRUEs,c(i,merged_pfocr$Symbol[j],as.numeric(table(response)['TRUE'])))
if(length(table(response))==1) next
roc_r=roc(response,as.numeric(rsea_results_human_voom_pfocr[[i]]$Comp.adjP))
auc_r=auc(roc_r)
if(auc_r<=0.5){rsea_pfocr_50lower=rbind(rsea_pfocr_50lower,c(i,j))}
auc_r_ci=ci(auc_r)
if(!is.na(auc_r_ci[1])){ if(auc_r_ci[1] < roc_r$auc[1] & auc_r_ci[2] > roc_r$auc[1]){
rsea_pfocr_aucs_ci=c(rsea_pfocr_aucs_ci,auc_r[1])
}}
rsea_pfocr_aucs=c(rsea_pfocr_aucs,auc_r[1])
}
}
boxplot(rsea_pfocr_aucs)
library(readxl)
library(org.Hs.eg.db)
data<-read_excel("/Users/mingyoungshin/git/PFOCRInPathwayAnalyses/human_knockout_genelist_alias_updated.xlsx", col_names = F)
data_gene=as.character(unlist(data[,2]))
gene_entrez <- bitr(data_gene, fromType = "SYMBOL",
toType = c("ENTREZID"),
OrgDb = org.Hs.eg.db, drop=FALSE)
colnames(gene_entrez)[1]="Gene"
gene_entrez=unique(gene_entrez)
tb=table(gene_entrez$Gene)
tb[which(tb>1)]
result_pfocr=matrix(NA,nrow=nrow(gene_entrez),ncol=3)
for(i in 1:nrow(result_pfocr)){
result_pfocr[i,1]=gene_entrez$Gene[i]
result_pfocr[i,2]=gene_entrez$ENTREZID[i]
for(j in 1:length(pfocr_3sets_list)){
temp=NULL
if( result_pfocr[i,2]%in%pfocr_3sets_list[[j]]){
temp=names(pfocr_3sets_list)[j]
}
if(length(temp)>0){result_pfocr[i,3]=paste(result_pfocr[i,3],temp,sep=";")}
}
}
result_pfocr=gsub("NA;","",result_pfocr)
nrow(result_pfocr)-sum(is.na(result_pfocr[,3]))
64
colnames(result_pfocr)=c("Symbol","Entrez","PFOCR")
colnames(data)=c("GSE","Symbol")
merged_pfocr=merge(data,result_pfocr,by="Symbol")
nrow(merged_pfocr)-sum(is.na(merged_pfocr[,3]))
85
write.table(merged_pfocr,"human_knockout_pfocr_3sets.txt",col.names = colnames(merged_pfocr),quote=FALSE)
library(readxl)
library(org.Hs.eg.db)
data<-read_excel("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/human_knockout_genelist_alias_updated_April2021.xlsx", col_names = F)
data_gene=as.character(unlist(data[,2]))
gene_entrez <- bitr(data_gene, fromType = "SYMBOL",
toType = c("ENTREZID"),
OrgDb = org.Hs.eg.db, drop=FALSE)
colnames(gene_entrez)[1]="Gene"
gene_entrez=unique(gene_entrez)
tb=table(gene_entrez$Gene)
tb[which(tb>1)]
result_pfocr=matrix(NA,nrow=nrow(gene_entrez),ncol=3)
for(i in 1:nrow(result_pfocr)){
result_pfocr[i,1]=gene_entrez$Gene[i]
result_pfocr[i,2]=gene_entrez$ENTREZID[i]
for(j in 1:length(pfocr_3sets_list)){
temp=NULL
if( result_pfocr[i,2]%in%pfocr_3sets_list[[j]]){
temp=names(pfocr_3sets_list)[j]
}
if(length(temp)>0){result_pfocr[i,3]=paste(result_pfocr[i,3],temp,sep=";")}
}
}
result_pfocr=gsub("NA;","",result_pfocr)
nrow(result_pfocr)-sum(is.na(result_pfocr[,3]))
40
colnames(result_pfocr)=c("Symbol","Entrez","PFOCR")
colnames(data)=c("GSE","Symbol")
merged_pfocr=merge(data,result_pfocr,by="Symbol")
nrow(merged_pfocr)-sum(is.na(merged_pfocr[,3]))
53
write.table(merged_pfocr,"human_knockout_pfocr_3sets_April2021version.txt",col.names = colnames(merged_pfocr),quote=FALSE)
rsea_results_human_voom_pfocr_3sets=list()
## process result.txt files for counts
for(i in 1:ncol(pvalue_results_human_voom)){
rsea_results_human_voom_pfocr_3sets[[i]]<-read.table(paste0("GSE/",colnames(pvalue_results_human_voom)[i],"/rSEA/PFOCR_3sets/result.txt"),header=TRUE)
}
merged_pfocr_3sets<-read.table("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/human_knockout_pfocr_3sets.txt",header=TRUE)
merged_pfocr_3sets<-read.table("/Users/mingyoungshin/Dropbox (Gladstone)/PFOCRInPathwayAnalyses_git/PFOCRInPathwayAnalyses/human_knockout_pfocr_3sets_April2021version.txt",header=TRUE)
targets=which(is.na(merged_pfocr_3sets$PFOCR)==FALSE)#get GSEs matched with knock genes
target_gses=merged_pfocr_3sets[targets,2]
length(target_gses)
target_gses=target_gses[-which(target_gses%in%c("GSE75291","GSE78506","GSE141126","GSE76421"))]
length(target_gses)
target_index=match(target_gses,colnames(pvalue_results_human_voom))#retrieve their index
target_index_pfocr_3sets=target_index
rsea_pfocr_3sets_sig_gene_count=NULL
rsea_pfocr_3sets_aucs=NULL
rsea_pfocr_3sets_aucs_ci=NULL
rsea_pfocr_3set_50lower=NULL
rsea_pfocr_3rsets_TRUEs=NULL
for(i in unique(target_index)){
print(i)
rsea_pfocr_3sets_sig_gene_count=c(rsea_pfocr_3sets_sig_gene_count,sum(rsea_results_human_voom_pfocr_3sets[[i]]$Com.adjP <0.05,na.rm=TRUE))
if( sum(is.na(rsea_results_human_voom_pfocr_3sets[[i]])) ==sum(lengths(rsea_results_human_voom_pfocr_3sets[[i]])) ){next}#pass null cases
gse_match=which(as.character(merged_pfocr_3sets$GSE)==colnames(pvalue_results_human_voom)[i])
for(j in gse_match){
print(merged_pfocr_3sets$PFOCR[j])
if(is.na(merged_pfocr_3sets$PFOCR[j])){next}
true_sig=merged_pfocr_3sets$PFOCR[j]
true_sig=as.character(strsplit(as.character(true_sig),";")[[1]])
true_sig=true_sig[true_sig%in%(names(pfocr_3sets_list))]
print(length(true_sig))
response=apply(as.matrix(1:length(rsea_results_human_voom_pfocr_3sets[[i]]$set_id)),1,function(x) rsea_results_human_voom_pfocr_3sets[[i]]$set_id[x]%in%true_sig)
print(sum(response))
print(table(response))
rsea_pfocr_3rsets_TRUEs=rbind(rsea_pfocr_3rsets_TRUEs,c(i,merged_pfocr_3sets$Symbol[j],as.numeric(table(response)['TRUE'])))
if(length(table(response))==1) next
roc_r=roc(response,as.numeric(rsea_results_human_voom_pfocr_3sets[[i]]$Comp.adjP))
auc_r=auc(roc_r)
if(auc_r<=0.5){rsea_pfocr_3set_50lower=rbind(rsea_pfocr_3set_50lower,c(i,j))}
auc_r_ci=ci(auc_r)
if(!is.na(auc_r_ci[1])){
if(auc_r_ci[1] < roc_r$auc[1] & auc_r_ci[2] > roc_r$auc[1]){
rsea_pfocr_3sets_aucs_ci=c(rsea_pfocr_3sets_aucs_ci,auc_r[1])
}
}
rsea_pfocr_3sets_aucs=c(rsea_pfocr_3sets_aucs,auc_r[1])
}
}
library(ggplot2)
data=rbind( cbind(rsea_pfocr_3sets_aucs,rep("PFOCR_3sets",length(rsea_pfocr_3sets_aucs))),
cbind(rsea_pfocr_aucs,rep("PFOCR",length(rsea_pfocr_aucs))),
cbind(rsea_wp_aucs,rep("WP",length(rsea_wp_aucs))),
cbind(rsea_go_aucs,rep("GO",length(rsea_go_aucs))) )
colnames(data)=c("AUC","Data")
data=data.frame(data)
data$AUC=as.numeric(as.character(data$AUC))
# Change violin plot colors by groups
p<-ggplot(data, aes(x=Data, y=AUC, fill=Data)) +
geom_boxplot() + scale_fill_manual(values=c("antiquewhite3", "darkseagreen", "darksalmon","lightblue")) + theme_classic()
p
data_ci=rbind( cbind(rsea_pfocr_3sets_aucs_ci,rep("PFOCR_3sets",length(rsea_pfocr_3sets_aucs_ci))),
cbind(rsea_pfocr_aucs_ci,rep("PFOCR",length(rsea_pfocr_aucs_ci))),
cbind(rsea_wp_aucs_ci,rep("WP",length(rsea_wp_aucs_ci))),
cbind(rsea_go_aucs_ci,rep("GO",length(rsea_go_aucs_ci))) )
colnames(data_ci)=c("AUC","Data")
data_ci=data.frame(data_ci)
data_ci$AUC=as.numeric(as.character(data_ci$AUC))
p<-ggplot(data_ci, aes(x=Data, y=AUC, fill=Data)) +
geom_boxplot() + scale_fill_manual(values=c("antiquewhite3", "darkseagreen", "darksalmon","lightblue")) + theme_classic()
p
saveRDS(data_ci,"data_ci.rds")
#################assess sets with auc <0.5
Reduce(intersect, list(rsea_go_50lower[,1],rsea_wp_50lower[,1],rsea_pfocr_3set_50lower[,1],rsea_pfocr_50lower[,1]))#0
library(ggvenn)
ggvenn(
list(rsea_go_50lower=rsea_go_50lower[,1],rsea_wp_50lower=rsea_wp_50lower[,1],rsea_pfocr_3set_50lower=rsea_pfocr_3set_50lower[,1],rsea_pfocr_50lower=rsea_pfocr_50lower[,1]),
fill_color = c("#0073C2FF", "#EFC000FF", "#868686FF", "#CD534CFF"),
stroke_size = 0.5, set_name_size = 4
)
table(c(rsea_go_50lower[,1],rsea_wp_50lower[,1],rsea_pfocr_3set_50lower[,1],rsea_pfocr_50lower[,1]))
#11 30 61 77 81 93 110 120 121 131 156 195 211 234 244 317 332
#1 2 3 2 1 1 2 1 3 3 2 1 1 1 2 2 1
#####################################################################################PCA plot
#################data 1
i=61
merged_filtered_gses[[i]]$GSE#GSE75215
geo_data<-GEOquery::getGEO(merged_filtered_gses[[i]]$GSE,getGPL = FALSE)
show(geo_data)
pheno=pData(geo_data[[1]])
dim(pheno)
colnames(pheno)
gsm_title=cbind(rownames(pheno),as.character(pheno$title))
head(gsm_title)
# Selected samples to be extracted
samp = gsm_title[,1]
# Identify columns to be extracted
sample_locations = which(samples %in% samp)
sample_locations
# extract gene expression from compressed data
expression = rhdf5::h5read(destination_file, "data/expression", index=list(1:length(genes), sample_locations))
rownames(expression) = genes
colnames(expression) = samples[sample_locations]
###DEG
filter <- apply(expression, 1, function(x) length(x[x>5])>=2) #minimum count (4) is satified in at least two conditions
FilterGeneCountData <- expression[filter,]
###group GSMs
Group=gsm_title[,2]
group_index=grep("group",names(merged_filtered_gses[[i]]))
selected_all=NULL
for(gi in group_index){
selected=which( (Group%in%unlist(merged_filtered_gses[[i]][gi])) ==TRUE)
Group[selected]=paste0("group",gi)
selected_all=c(selected_all,selected)
}
gsm_title1=cbind(gsm_title,Group)
gsm_title1=gsm_title1[selected_all,]
colnames(gsm_title1)=c("GSM","Pheno","Group")
gsm_title1=as.data.frame(gsm_title1)
gsm_title1$Group <- as.factor(gsm_title1$Group)
TempIndices <- match(gsm_title1[,1], colnames(FilterGeneCountData))
colN=colnames(FilterGeneCountData)
FilterGeneCountData=FilterGeneCountData[,TempIndices]
y <- DGEList(counts=FilterGeneCountData, group=NULL)
y <- calcNormFactors(y)
cpm <- cpm(y, log = TRUE, prior.count = 0.01)
rv <- apply(cpm,1,var)
#Select genes with highest variance.
keep <- order(rv, decreasing = TRUE)[1:500]
selected <- cpm[keep, ] %>% t()
#Transpose is needed to ensure that each row is a vector.
pca <- prcomp(selected, scale=T, center = T)
stddev <- pca$sdev
pc1_var <- round(100*stddev[1]^2/sum(stddev^2))
pc2_var <- round(100*stddev[2]^2/sum(stddev^2))
pc3_var <- round(100*stddev[3]^2/sum(stddev^2))
PlotData <- data.frame(cbind(PC1 = pca$x[,1], PC2 = pca$x[,2],PC3 = pca$x[,3]))
PlotData <- gsm_title1[, c("Group")] %>%
cbind(PlotData, .)
colnames(PlotData)[4]="Group"
gp1=ggplot(PlotData, aes(x=PC1, y=PC2, color=Group, shape=Group)) +
geom_point(size=4.5) +
xlab(paste("PC1:", pc1_var, "% variance")) +
ylab(paste("PC2:", pc2_var, "% variance"))
gp1
jpeg(paste0("PCA_",merged_filtered_gses[[i]]$GSE,"_index",i,".jpeg"))
print(gp1)
dev.off()
#######################
GSE68925 ROCK1
GSE75215 PRMT5
GSE59598 ZNF423
GSE99596 SIRT7
GSE96637 ROR1
GSE76421 JAG1
GSE108500 FOXP1
GSE77140 TRIM24
GSE118532 RNF168
#######################check true gene set count
tb=table(c(rsea_go_50lower[,1],rsea_wp_50lower[,1],rsea_pfocr_3set_50lower[,1],rsea_pfocr_50lower[,1]))
tb=cbind(names(tb),tb)
colnames(tb)=c("GSE_index","Count_database_low_performance")
colnames(rsea_pfocr_3rsets_TRUEs)=c("GSE_index","knockgene","Count_True_geneset")
colnames(rsea_pfocr_TRUEs)=c("GSE_index","knockgene","Count_True_geneset")
colnames(rsea_wp_TRUEs)=c("GSE_index","knockgene","Count_True_geneset")
colnames(rsea_go_TRUEs)=c("GSE_index","knockgene","Count_True_geneset")
rsea_pfocr_3rsets_TRUEs_merged=merge(rsea_pfocr_3rsets_TRUEs,tb,by="GSE_index")
rsea_pfocr_3rsets_TRUEs_merged=rsea_pfocr_3rsets_TRUEs_merged[order(rsea_pfocr_3rsets_TRUEs_merged[,4]),]
rsea_pfocr_TRUEs_merged=merge(rsea_pfocr_TRUEs,tb,by="GSE_index")
rsea_pfocr_TRUEs_merged=rsea_pfocr_TRUEs_merged[order(rsea_pfocr_TRUEs_merged[,4]),]
rsea_wp_TRUEs_merged=merge(rsea_wp_TRUEs,tb,by="GSE_index")
rsea_wp_TRUEs_merged=rsea_wp_TRUEs_merged[order(rsea_wp_TRUEs_merged[,4]),]
rsea_go_TRUEs_merged=merge(rsea_go_TRUEs,tb,by="GSE_index")
rsea_go_TRUEs_merged=rsea_go_TRUEs_merged[order(rsea_go_TRUEs_merged[,4]),]
save(rsea_pfocr_3rsets_TRUEs_merged,rsea_pfocr_TRUEs_merged,rsea_wp_TRUEs_merged,rsea_go_TRUEs_merged,file="rsea_GSE_knockoutGene_TRUEgeneset_lowPerformanceDBcount.RData")
permute_databases.sh
ob - mean(null)/sd(null)