-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path08_GSEA_rlog-as-Rank.R
156 lines (133 loc) · 4.87 KB
/
08_GSEA_rlog-as-Rank.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
library(ComplexHeatmap)
library(corrplot)
library(circlize)
library(fgsea)
library(data.table)
library(clusterProfiler)
library(ggplot2)
library(fastDummies)
library(tidyr)
library(dplyr)
options(stringsAsFactors = F)
# ------------------- Function Block ------------------- #
read.gmt <- function(fname){
res <- list(genes=list(),
desc=list())
gmt <- file(fname)
gmt.lines <- readLines(gmt)
close(gmt)
gmt.list <- lapply(gmt.lines,
function(x) unlist(strsplit(x, split="\t")))
gmt.names <- sapply(gmt.list, '[', 1)
gmt.desc <- lapply(gmt.list, '[', 2)
gmt.genes <- lapply(gmt.list,
function(x){x[3:length(x)]})
names(gmt.desc) <- names(gmt.genes) <- gmt.names
return(gmt.genes)
}
ssGSEA <- function(gmtfile=gmtfile,fileranks=fileranks,Ptype=Ptype,pval_cutoff=pval_cutoff){
gmt <- read.gmt(gmtfile)
ranks_df <- read.delim(fileranks)
names <- colnames(ranks_df)
ranks_df$geneID <- row.names(ranks_df)
ranks_df <- ranks_df %>% select(geneID, all_of(names))
ranks_df <- ranks_df[complete.cases(ranks_df), ]
ranks_ch <- colnames(ranks_df)[-1]
ranks_ch <- setNames(ranks_ch, ranks_ch)
#run fastGSEA
tmp_ranks <- lapply(ranks_ch, function(rankname){
tmpdf <- ranks_df[,c('geneID', rankname)]
tmpdf <- tmpdf[complete.cases(tmpdf),]
tmpranks <- tmpdf[[rankname]]
names(tmpranks) <- tmpdf$geneID
fgseaRes <- fgsea(pathways = gmt, stats = tmpranks,
minSize = 15, maxSize = 2000, nperm = 1000)
fgseaRes <- as.data.frame(fgseaRes)
fgseaRes <- fgseaRes[,c('pathway', 'pval', 'padj', 'NES','size','leadingEdge')]
fgseaRes
})
# Remove ranks without enrichment
tmp_ranks <- Filter(function(x) nrow(x) > 1, tmp_ranks)
# Write output NES
rm(df)
for(name_rank in names(tmp_ranks)){
rank_out <- tmp_ranks[[name_rank]]
xxx <- rank_out[,"NES"]
names(xxx) <- rank_out[,"pathway"]
df <- cbind(df,xxx)
colnames(df)[ncol(df)] <- name_rank
}
df <- df[,-1]
rank_nameout <- paste0('NES_', fileranks,sep="")
write.table(df, file = rank_nameout, sep = '\t', quote = F, row.names = T,col.names=NA)
# Write output AdjP
rm(df)
for(name_rank in names(tmp_ranks)){
rank_out <- tmp_ranks[[name_rank]]
xxx <- rank_out[,"padj"]
names(xxx) <- rank_out[,"pathway"]
df <- cbind(df,xxx)
colnames(df)[ncol(df)] <- name_rank
}
df <- df[,-1]
rank_nameout <- paste0('padj_', fileranks,sep="")
write.table(df, file = rank_nameout, sep = '\t', quote = F, row.names = T,col.names=NA)
# Write output pval
rm(df)
for(name_rank in names(tmp_ranks)){
rank_out <- tmp_ranks[[name_rank]]
xxx <- rank_out[,"pval"]
names(xxx) <- rank_out[,"pathway"]
df <- cbind(df,xxx)
colnames(df)[ncol(df)] <- name_rank
}
df <- df[,-1]
rank_nameout <- paste0('pval_', fileranks,sep="")
write.table(df, file = rank_nameout, sep = '\t', quote = F, row.names = T,col.names=NA)
# Write output Leading Edge genes
rm(df)
for(name_rank in names(tmp_ranks)){
rank_out <- tmp_ranks[[name_rank]]
xxx <- rank_out[,"leadingEdge"]
names(xxx) <- rank_out[,"pathway"]
df <- cbind(df,xxx)
colnames(df)[ncol(df)] <- name_rank
}
df <- df[,-1]
rank_nameout <- paste0('LE_', fileranks,sep="")
write.table(df, file = rank_nameout, sep = '\t', quote = F, row.names = T,col.names=NA)
nes <- data.table::fread(paste0('NES_', fileranks,sep=""))
colnames(nes)[1] <- "pathway"
pval <- data.table::fread(paste0(Ptype,'_', fileranks,sep=""))
colnames(pval)[1] <- "pathway"
nes_melt <- nes %>%
gather(sample, nes, -pathway)
pval_melt <- pval %>%
gather(sample, pval, -pathway)
result <- full_join(nes_melt, pval_melt, by=c("pathway", "sample")) %>%
filter(pval <= pval_cutoff) %>%
select(pathway, sample, nes) %>%
spread(sample, nes)
rank_nameout <- paste0('NES_',Ptype,pval_cutoff,"_", fileranks,sep="")
write.table(result, file = rank_nameout, sep = '\t', quote = F, row.names = T,col.names=NA)
}
# ------------------- End of function block ------------------- #
### This steep is important to perform Spearman correlation between the NES (interferon signaling pathway)
### and cell type scores for each children.
# ----------- Prepare input for GSEA ------------------- #
# Preparing rlog table to be used as a rank
rlogrank <- read.delim("data/processed/Baseline_nasal-2018_selected_children_rlogNorm.tsv")
kidslist <- colnames(rlogrank)
rlogrank$means <- rowMeans(rlogrank)
rlogrank <- rlogrank %>% arrange(means) %>% select(all_of(kidslist))
write.table(rlogrank, "data/GSEA/RlogRanks_baseline.tsv", sep ="\t", quote = F)
setwd("data/GSEA/")
fileranks <- "RlogRanks_baseline.tsv"
gmtfile <- "ReactomePathwaysLevel3.gmt"
Ptype <- "padj"
pval_cutoff <- 0.25
#Run ssGSEA
ssGSEA(gmtfile = gmtfile,
fileranks = fileranks,
Ptype = Ptype,
pval_cutoff = pval_cutoff)