-
Notifications
You must be signed in to change notification settings - Fork 0
/
callCNV.R
168 lines (140 loc) · 6.08 KB
/
callCNV.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
#!/usr/bin/env Rscript
require(optparse)
# parse options ----------------------------------------------------------
# Inputs:
option_list = list(
make_option(c("-c", "--ctrl"), type="character", default="./data/hg38_75bp/GRCh38_75.avg_cov.w1000K_s500K.bed",
help="ctrl bed file [default= %default]", metavar="character"),
make_option(c("-t", "--treat"), type="character", default="out.txt",
help="treat bed file", metavar="character"),
make_option(c("-g", "--genome"), type="character", default="../data/hg38.chrom.sizes",
help="genome size file [default= %default]", metavar="character"),
make_option(c("-s", "--sample"), type="character", default=NULL,
help="sample name", metavar="character"),
make_option(c("-u", "--upper"), type="double", default=2,
help="upper boundary in trimming log2Ratio [default= %default]", metavar="double"),
make_option(c("-m", "--mapb_th"), type="double", default=0.75,
help="mapbility threshold for regions [default= %default]", metavar="double"),
make_option(c("-o", "--output_dir"), type="character", default="./data/output/",
help="output dir [default= %default]", metavar="character"),
make_option(c("-l", "--lower"), type="double", default=-1.5,
help="lower boundary in trimming log2Ratio [default= %default]", metavar="double")
);
opt_parser = OptionParser(option_list=option_list);
opt = parse_args(opt_parser);
# check if treat exist
if (!file.exists(opt$treat)){
print_help(opt_parser)
stop("(treat bed file not exist!).n", call.=FALSE)
}
if (!file.exists(opt$ctrl)){
print_help(opt_parser)
stop("(control bed file not exist!).n", call.=FALSE)
}
if (is.null(opt$sample)){
opt$sample = sub(".bed","",sub(".*/","",opt$treat))
}
# Intro ------------------------------------------------------------------
suppressMessages(source("./ultility.R"))
# ctrl
dat.ctrl <- import(opt$ctrl,format = "bed",genome=opt$genome)
dat.ctrl <- filter_gr(dat.ctrl)
# treat
dat.treat <- import(opt$treat,format = "bed",genome=hg38.chrom.sizes)
dat.treat <- filter_gr(dat.treat)
# Check ctrl will be used to weight the coverage at each location
if (!all.equal(length(match(dat.ctrl,dat.treat)), length(dat.ctrl), length(dat.treat))){
print_help(opt_parser)
stop("(contrl and treat are not the same resolution!).n", call.=FALSE)
}
# assign median to low mapb regions
idx.filter <- score(dat.ctrl)>=opt$mapb_th
dat.treat.mod <- dat.treat
score(dat.treat.mod)[!idx.filter] <- median(score(dat.treat.mod)[idx.filter])
raw.t <- score(dat.treat.mod)/median(score(dat.treat.mod))
summary(raw.t)
raw.c <- score(dat.ctrl)
raw.c[!idx.filter] <- median(raw.c[idx.filter])
# standardizatin & scaling
ratio.t <- raw.t/(raw.c+1)*2 # adjusted coverage by mapb
ratio.t <- ratio.t/median(ratio.t) # ratio of sample's median
score(dat.treat.mod) <- ratio.t
ratio.t <- score(scale_by_chr(dat.treat.mod)) # scaled by each chr's median
ratio.t.log2 <- log2(ratio.t) # convert to log2Ratio
# (optional) trim the log2Ratio (upper, lower )
ratio.t.log2[ratio.t.log2> opt$upper] <- opt$upper
ratio.t.log2[ratio.t.log2 < opt$lower] <- opt$lower
# create CNA object
CNA.object <- CNA(cbind(ratio.t.log2),
as.character(seqnames(dat.treat.mod)),
start(dat.treat.mod),
data.type="logratio",sampleid=opt$sample)
# smooth track by removing outliner and smooth to neighbours
smoothed.CNA.object <- smooth.CNA(CNA.object)
# detect segments
#segment.smoothed.CNA.object <- segment(smoothed.CNA.object, verbose=1)
sdundo.CNA.object <- segment(smoothed.CNA.object,
undo.splits="sdundo",
undo.SD=3,verbose=1)
# plots
if(F){
pdf(file=paste0(opt$output_dir,opt$sample,".pdf"),width = 12,height = 7)
plot(segment.smoothed.CNA.object, plot.type="w",ylim=c(-2,2))
plot(segment.smoothed.CNA.object, plot.type="s",ylim=c(-2,2))
plot(segment.smoothed.CNA.object, plot.type="p",ylim=c(-2,2))
dev.off()}
pdf(file=paste0(opt$output_dir,opt$sample,"_SD.pdf"),width = 12,height = 7)
plot(sdundo.CNA.object, plot.type="w",ylim=c(-2,2))
plot(sdundo.CNA.object, plot.type="s",ylim=c(-2,2))
plot(sdundo.CNA.object, plot.type="p",ylim=c(-2,2))
dev.off()
# export segments by chr and ranking
write_rds(list(opt=opt,
smoothed.CNA.object =smoothed.CNA.object,
sdundo.CNA.object=sdundo.CNA.object),
path = paste0(opt$output_dir,opt$sample,".rds"))
write_csv(sdundo.CNA.object$output,
path = paste0(opt$output_dir,opt$sample,"_SD.csv"))
if(sum(abs(sdundo.CNA.object$output$seg.mean)>=1)>0)
write.bed.seg(sdundo.CNA.object$output%>% filter(abs(seg.mean)>=1),
fn = paste0(opt$output_dir,opt$sample,".cnv.bed"))
# optional cmds -----------------------------------------------------------
#glFrequency(segment.smoothed.CNA.object)
#plotSample(sdundo.CNA.object,ylim=c(-3,3))
#zoomIntoRegion(sdundo.CNA.object, chrom="chr15",
# maploc.start = 57851725,maploc.end = 64357541,
# sampleid="GBM3",
# pt.pch = 16,pt.cex = .5,ylim=c(-3,3))
#abline(h=0,col='grey')
if(F){
zoomIntoRegion(sdundo.CNA.object, chrom="chr8",
#maploc.start = 67068340,maploc.end = 86000925,
sampleid="GBM3",
pt.pch = 16,pt.cex = .5,ylim=c(-2,2))
abline(h=c(-1,1),col='grey')
}
if(F){
png(file="./atacCNV/figs/GBM3_cmp_smooth.png")
par(mfrow=c(2,1))
plot(CNA.object$GBM3,ylim=c(-2.5,8),cex=0.25)
plot(smoothed.CNA.object$GBM3,ylim=c(-2.5,8),cex=0.25)
par(mfrow=c(1,1))
dev.off()
}
### case 1: mapb = 0
if(F){
hist(score(dat.ctrl),nclass = 50)
summary(score(dat.ctrl))
summary(score(dat.treat))
#plot(CNA.object, plot.type="")
plot(segment.smoothed.CNA.object, plot.type="p")
plot(segment.smoothed.CNA.object, plot.type="w")
plot(sdundo.CNA.object,plot.type="p")
plot(sdundo.CNA.object,plot.type="w")
plot(sdundo.CNA.object,plot.type="s")
# check segmentation
par(mfrow=c(2,1))
plot(segment.smoothed.CNA.object, plot.type="w")
plot(sdundo.CNA.object,plot.type="w")
par(mfrow=c(1,1))
}