-
Notifications
You must be signed in to change notification settings - Fork 9
/
DESeq2-example.R
197 lines (152 loc) · 8.7 KB
/
DESeq2-example.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
library(DESeq2)
library(RColorBrewer)
library(gplots)
source("https://bioconductor.org/biocLite.R")
biocLite("biomaRt", suppressUpdates = TRUE)
library(biomaRt)
library(data.table)
library(stringr)
# convert ENSG.# genes to ENSG, then gene symbols----
setwd("Countfiles_ENSG") # htseq count file output from Chromium data comes with ENSG gene IDs
files <- Sys.glob("Sample*counts.txt")
for (file in files){
dt <- fread(file)
dt <- dt[1:(nrow(dt)-5),]
names(dt) <- c("ENSG.full", "Counts")
ensg.genes <- data.table("ENSG.full" = dt$ENSG.full)
ensg.genes[,ENSG.short := tstrsplit(ENSG.full, "\\.")[1]]
genes <- ensg.genes$ENSG.short
mart <- useDataset("hsapiens_gene_ensembl", useMart("ensembl"))
G_list <- getBM(filters= "ensembl_gene_id", attributes= c("ensembl_gene_id","external_gene_name"),values=genes,mart= mart)
G_list <- as.data.table(G_list)
ensg.genes <- merge(ensg.genes, G_list, all=T, by.x="ENSG.short", by.y="ensembl_gene_id")
ensg.genes <- na.omit(ensg.genes)
dt <- subset(dt, dt$ENSG.full %in% ensg.genes$ENSG.full)
dt <- merge(dt, ensg.genes, by="ENSG.full")
dt <- dt[,c("external_gene_name", "Counts")]
dt <- dt[!duplicated(external_gene_name),]
sample_id <- str_split_fixed(file, "_htseq_counts.txt",2)[1]
fwrite(dt, paste0("Countfiles_gene_symbols/", sample_id, "_gene_symbol_counts.txt"), col.names = F, row.names = F, quote=F, sep="\t")
}
# start deseq2 work----
sampleTable <- fread("sampletable.csv")
sampleFiles <- gtools::mixedsort(grep("Sample_",list.files("./Countfiles_gene_symbols/"),value=TRUE))
sampleCondition <- unlist(sampleTable[,condition])
sampleTable <- data.frame(sampleName = sampleFiles,
fileName = sampleFiles,
condition = sampleCondition)
sampleTable
directory <- "Countfiles_gene_symbols/"
dds <- DESeqDataSetFromHTSeqCount(sampleTable = sampleTable,
directory = directory,
design = ~ condition)
keep <- rowSums(counts(dds)) >= 10
dds <- dds[keep,]
dds <- DESeq(dds)
# DE TESTS----
res <- results(dds)
res.day0.vs.day4 <- lfcShrink(dds, contrast=c("condition","H9noc_D4", "H9noc_D0")) # default padjust method is Benjamini-Hochberg. must specify "later timepoint" versus "early timepoint" to be in the same log fold change direction as "early versus later" in results.
res.day0.vs.day8 <- lfcShrink(dds, contrast=c("condition","H9noc_D8", "H9noc_D0"))
res.day0.vs.day12 <- lfcShrink(dds, contrast=c("condition","H9noc_D12", "H9noc_D0"))
res.day0.vs.day21 <- lfcShrink(dds, contrast=c("condition","H9noc_D21", "H9noc_D0"))
res.day0.vs.day28 <- lfcShrink(dds, contrast=c("condition","H9noc_D28", "H9noc_D0"))
res.day4.vs.day8 <- lfcShrink(dds, contrast=c("condition","H9noc_D8", "H9noc_D4"))
res.day4.vs.day12 <- lfcShrink(dds, contrast=c("condition","H9noc_D12", "H9noc_D4"))
res.day4.vs.day21 <- lfcShrink(dds, contrast=c("condition","H9noc_D21", "H9noc_D4"))
res.day4.vs.day28 <- lfcShrink(dds, contrast=c("condition","H9noc_D28", "H9noc_D4"))
res.day8.vs.day12 <- lfcShrink(dds, contrast=c("condition","H9noc_D12", "H9noc_D8"))
res.day8.vs.day21 <- lfcShrink(dds, contrast=c("condition","H9noc_D21", "H9noc_D8"))
res.day8.vs.day28 <- lfcShrink(dds, contrast=c("condition","H9noc_D28", "H9noc_D8"))
res.day12.vs.day21 <- lfcShrink(dds, contrast=c("condition","H9noc_D21", "H9noc_D12"))
res.day12.vs.day28 <- lfcShrink(dds, contrast=c("condition","H9noc_D28", "H9noc_D12"))
res.day21.vs.day28 <- lfcShrink(dds, contrast=c("condition","H9noc_D28", "H9noc_D21"))
setwd(directory)
save(dds, res.day0.vs.day12, res.day0.vs.day21, res.day0.vs.day28, res.day0.vs.day4, res.day0.vs.day8, res.day12.vs.day21, res.day12.vs.day28, res.day21.vs.day28, res.day4.vs.day12, res.day4.vs.day21, res.day4.vs.day28, res.day4.vs.day8, res.day8.vs.day12, res.day8.vs.day21, res.day8.vs.day28, sampleTable, file="Tao_nociceptor_DDS.Rdata")
#MA plots----
plotMA(res.day0.vs.day28, alpha=(1*10^(-5)))
#reformat to data.table and add mean per level----
reformat.res <- function(x, condition1, condition2){
x <- as.data.frame(x)
x$GeneId <- row.names(x)
x <- as.data.table(x)
x[,baseMean:=NULL]
x[,baseMean1 := rowMeans(counts(dds,normalized=TRUE)[,dds$condition == condition1])]
x[,baseMean2 := rowMeans(counts(dds,normalized=TRUE)[,dds$condition == condition2])]
x <- na.omit(x)
x <- x[order(rank(padj))]
x <- x[,c("GeneId", "baseMean1", "baseMean2", "log2FoldChange", "lfcSE", "stat", "pvalue", "padj")]
names(x)[2:3] <- c(condition1, condition2)
return(x)
}
res.day0.vs.day4 <- reformat.res(res.day0.vs.day4, "H9noc_D0", "H9noc_D4")
res.day0.vs.day8 <- reformat.res(res.day0.vs.day8, "H9noc_D0", "H9noc_D8")
res.day0.vs.day12 <- reformat.res(res.day0.vs.day12, "H9noc_D0", "H9noc_D12")
res.day0.vs.day21 <- reformat.res(res.day0.vs.day21, "H9noc_D0", "H9noc_D21")
res.day0.vs.day28 <- reformat.res(res.day0.vs.day28, "H9noc_D0", "H9noc_D28")
res.day4.vs.day8 <- reformat.res(res.day4.vs.day8, "H9noc_D4", "H9noc_D8")
res.day4.vs.day12 <- reformat.res(res.day4.vs.day12, "H9noc_D4", "H9noc_D12")
res.day4.vs.day21 <- reformat.res(res.day4.vs.day21, "H9noc_D4", "H9noc_D21")
res.day4.vs.day28 <- reformat.res(res.day4.vs.day28, "H9noc_D4", "H9noc_D28")
res.day8.vs.day12 <- reformat.res(res.day8.vs.day12, "H9noc_D8", "H9noc_D12")
res.day8.vs.day21 <- reformat.res(res.day8.vs.day21, "H9noc_D8", "H9noc_D21")
res.day8.vs.day28 <- reformat.res(res.day8.vs.day28, "H9noc_D8", "H9noc_D28")
res.day12.vs.day21 <- reformat.res(res.day12.vs.day21, "H9noc_D12", "H9noc_D21")
res.day12.vs.day28 <- reformat.res(res.day12.vs.day28, "H9noc_D12", "H9noc_D28")
res.day21.vs.day28 <- reformat.res(res.day21.vs.day28, "H9noc_D21", "H9noc_D28")
# write----
setwd("DE")
fwrite(res.day0.vs.day4, "DE.H9noc_D0_vs_D4.csv") #
fwrite(res.day0.vs.day8, "DE.H9noc_D0_vs_D8.csv")
fwrite(res.day0.vs.day12, "DE.H9noc_D0_vs_D12.csv")
fwrite(res.day0.vs.day21, "DE.H9noc_D0_vs_D21.csv")
fwrite(res.day0.vs.day28, "DE.H9noc_D0_vs_D28.csv")
fwrite(res.day4.vs.day8, "DE.H9noc_D4_vs_D8.csv")#
fwrite(res.day4.vs.day12, "DE.H9noc_D4_vs_D12.csv")
fwrite(res.day4.vs.day21, "DE.H9noc_D4_vs_D21.csv")
fwrite(res.day4.vs.day28, "DE.H9noc_D4_vs_D28.csv")
fwrite(res.day8.vs.day12, "DE.H9noc_D8_vs_D12.csv")#
fwrite(res.day8.vs.day21, "DE.H9noc_D8_vs_D21.csv")
fwrite(res.day8.vs.day28, "DE.H9noc_D8_vs_D28.csv")
fwrite(res.day12.vs.day21, "DE.H9noc_D12_vs_D21.csv")#
fwrite(res.day21.vs.day28, "DE.H9noc_D21_vs_D28.csv")#
#volcano plot----
#restoplot <- res.day0.vs.day4 # used 3 for lfc threshold
#restoplot <- res.day4.vs.day8 # 6
#restoplot <- res.day8.vs.day12 # 5
#restoplot <- res.day0.vs.day28
#restoplot <- res.day12.vs.day21 #5
restoplot <- res.day21.vs.day28
alpha <- 1*10^(-50) # Threshold on the adjusted p-value
cols <- densCols(restoplot$log2FoldChange, -log10(restoplot$pvalue))
plot(restoplot$log2FoldChange, -log10(restoplot$padj), col=cols, panel.first=grid(),
main="Volcano plot: D21 vs D28", xlab="Effect size: log2(fold-change)", ylab="-log10(adjusted p-value)",
pch=20, cex=0.6, xlim=c(-10,10))
abline(v=0)
abline(v=c(-1,1), col="brown")
abline(h=-log10(alpha), col="brown")
gn.selected <- abs(restoplot$log2FoldChange) > 4 & restoplot$padj < alpha
text(restoplot$log2FoldChange[gn.selected],
-log10(restoplot$padj)[gn.selected],
lab=(restoplot$GeneId)[gn.selected ], cex=0.7)
# condition1 higher is on the left, condition2 higher is on the right
# for plotting counts of a particular gene-----
plotCounts(dds, gene=which.min(res$padj), intgroup="condition")
plotCounts(dds, gene="HECW1", intgroup="condition")
# prepare for GO tests----
setwd("GO/")
filtered <- res.day0.vs.day4[abs(log2FoldChange)>1 & padj < 0.001 & H9noc_D0 > 25 & H9noc_D4 > 25,] # filtered set.
filtered <- as.data.table(filtered$GeneId[1:200])
fwrite(filtered, "H9noc_D0_vs_D4.txt", col.names = F, row.names = F)
filtered <- res.day4.vs.day8[abs(log2FoldChange)>1 & padj < 0.001 & H9noc_D4 > 25 & H9noc_D8 > 25,] # filtered set.
filtered <- as.data.table(filtered$GeneId[1:200])
fwrite(filtered, "H9noc_D4_vs_D8.txt", col.names = F, row.names = F)
filtered <- res.day8.vs.day12[abs(log2FoldChange)>1 & padj < 0.001 & H9noc_D8 > 25 & H9noc_D12 > 25,] # filtered set.
filtered <- as.data.table(filtered$GeneId[1:200])
fwrite(filtered, "H9noc_D8_vs_D12.txt", col.names = F, row.names = F)
filtered <- res.day12.vs.day21[abs(log2FoldChange)>1 & padj < 0.001 & H9noc_D12 > 25 & H9noc_D21 > 25,] # filtered set.
filtered <- as.data.table(filtered$GeneId[1:200])
fwrite(filtered, "H9noc_D12_vs_D21.txt", col.names = F, row.names = F)
filtered <- res.day21.vs.day28[abs(log2FoldChange)>1 & padj < 0.001 & H9noc_D21 > 25 & H9noc_D28 > 25,] # filtered set.
filtered <- as.data.table(filtered$GeneId[1:200])
fwrite(filtered, "H9noc_D21_vs_D28.txt", col.names = F, row.names = F)
#