/
SNP-calling.Rmd
306 lines (235 loc) · 14.5 KB
/
SNP-calling.Rmd
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
---
title: "### SNP-calling"
author: "Ruijuan Li"
date: "5/7/2018"
output:
html_document:
keep_md: yes
---
### Freebayes SNP calling
```{r}
# 1) extract uniquely mapped reads
# samtools view Aligned.sortedByCoord.out.bam | awk '$5 == "255"' > Ae_no_bolting_unique_mapped.sam
# samtools view Aligned.sortedByCoord.out.bam | awk '$5 == "255"' > Ol_no_bolting_unique_mapped.sam
# 2) from sam to bam
# samtools view -bT Brassica_napus_v4.1.chromosomes.fa Ae_no_bolting_unique_mapped.sam > Ae_no_bolting_unique_mapped.bam
# samtools view -bT Brassica_napus_v4.1.chromosomes.fa Ol_no_bolting_unique_mapped.sam > Ol_no_bolting_unique_mapped.bam
# 3) remove PCR duplicate
# samtools rmdup -s Ae_unique_sorted.bam Ae_no_bolting_unique_mapped_rmdup.bam
# samtools rmdup -s Ol_unique_sorted.bam Ol_no_bolting_unique_mapped_rmdup.bam
# 4) index bam file
# samtools index Ae_no_bolting_unique_mapped_rmdup.bam
# samtools index Ol_no_bolting_unique_mapped_rmdup.bam
# 5) add read group and SNP calling
# freebayes_v0.9.21-7-g7dd41db -f /share/malooflab/John/KIAT/Reference/Brassica_napus_v4.1.chromosomes.fa Ae_no_bolting_unique_mapped_rmdup.bam --genotype-qualities --dont-left-align-indels > Ae.vcf
# freebayes_v0.9.21-7-g7dd41db -f /share/malooflab/John/KIAT/Reference/Brassica_napus_v4.1.chromosomes.fa Ol_no_bolting_unique_mapped_rmdup.bam --genotype-qualities --dont-left-align-indels > Ol.vcf
# 6) remove sam files
# rm Ae_no_bolting_unique_mapped.sam
# rm Ol_no_bolting_unique_mapped.sam
```
### GATK SNP calling
```{r}
# 1) sort
# samtools sort Ae_no_bolting_unique_mapped.bam -o Ae_unique_sorted.bam
# samtools sort Ol_no_bolting_unique_mapped.bam -o Ol_unique_sorted.bam
# 2) add readgroup
# picard AddOrReplaceReadGroups I=Ae_unique_sorted.bam O=Ae_unique_sorted_addrg.bam RGID=Ae RGLB=lib1 RGPL=illumina RGPU=unit1 RGSM=Ae
# picard AddOrReplaceReadGroups I=Ol_unique_sorted.bam O=Ol_unique_sorted_addrg.bam RGID=Ol RGLB=lib1 RGPL=illumina RGPU=unit1 RGSM=Ol
# 3) mark PCR duplicate
# picard MarkDuplicates I=Ae_unique_sorted_addrg.bam O=Ae_unique_sorted_addrg_marked_duplicates.bam M=marked_dup_metrics_Ae.txt
# picard MarkDuplicates I=Ol_unique_sorted_addrg.bam O=Ol_unique_sorted_addrg_marked_duplicates.bam M=marked_dup_metrics_Ol.txt
# 4) index bam file
# samtools index Ae_unique_sorted_addrg_marked_duplicates.bam
# samtools index Ol_unique_sorted_addrg_marked_duplicates.bam
# 5) create fastq sequence dictionary file (only need to do once, under the ref fa file folder)
# picard CreateSequenceDictionary R=Brassica_napus_v4.1.chromosomes.fa O=Brassica_napus_v4.1.chromosomes.dict
# 6) split 'N' trim (visualize bam alignment before & after alignment in IGV)
# GATK -T SplitNCigarReads -R Brassica_napus_v4.1.chromosomes.fa -I Ae_unique_sorted_addrg_marked_duplicates.bam -o Ae_unique_sorted_addrg_marked_duplicates_split.bam -rf ReassignOneMappingQuality -RMQF 255 -RMQT 60 -U ALLOW_N_CIGAR_READS
# GATK -T SplitNCigarReads -R Brassica_napus_v4.1.chromosomes.fa -I Ol_unique_sorted_addrg_marked_duplicates.bam -o Ol_unique_sorted_addrg_marked_duplicates_split.bam -rf ReassignOneMappingQuality -RMQF 255 -RMQT 60 -U ALLOW_N_CIGAR_READS
# 7) determine suspicious intervals which are likely in need of realignment
# GATK -T RealignerTargetCreator -R Brassica_napus_v4.1.chromosomes.fa -I Ae_unique_sorted_addrg_marked_duplicates_split.bam -o Ae_forIndelRealigner.intervals
# GATK -T RealignerTargetCreator -R Brassica_napus_v4.1.chromosomes.fa -I Ol_unique_sorted_addrg_marked_duplicates_split.bam -o Ol_forIndelRealigner.intervals
# 8) running realigner over those intervals
# GATK -T IndelRealigner -R Brassica_napus_v4.1.chromosomes.fa -I Ae_unique_sorted_addrg_marked_duplicates_split.bam -targetIntervals Ae_forIndelRealigner.intervals -o Ae_realignedBam.bam
# GATK -T IndelRealigner -R Brassica_napus_v4.1.chromosomes.fa -I Ol_unique_sorted_addrg_marked_duplicates_split.bam -targetIntervals Ol_forIndelRealigner.intervals -o Ol_realignedBam.bam
# 9) variant calling through slurm schedular using script https://github.com/MaloofLab/Li-eQTL-2018/blob/master/scripts/SNP_calling_GATK.slurm
```
### SNP filtering for Freebayes output
```{r}
library(boxr) # need to put data to box for loading
source("~/Desktop/F2_paper/submission/Li-eQTL-2018/scripts/helper.R")
# use Da-Ae as an example for SNP filtering from Freebayes
vcf.freebayes.Ae <- read.table("~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/data/freebayes/as_diploid/no_bolting_unique_mapped/Ae_SNP_biallelic.recode.vcf",as.is=T,na.strings = ".:.:.:.:.:.:.")
vcf.header.freebayes.Ae <- system("grep '#C' ~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/data/freebayes/as_diploid/no_bolting_unique_mapped/Ae_SNP_biallelic.recode.vcf",intern = TRUE)
vcf.header.freebayes.Ae <- sub("#","",vcf.header.freebayes.Ae)
vcf.header.freebayes.Ae <- unlist(strsplit(vcf.header.freebayes.Ae,split="\t"))
vcf.freebayes.reform.Ae <- SNP.freebayes.reformat.Ae(vcf.freebayes.Ae, vcf.header.freebayes.Ae)
# QUAL > 40
vcf.freebayes.HQ.Ae <- vcf.freebayes.reform.Ae[vcf.freebayes.reform.Ae$QUAL>40,]
# 10 < depth < 1000
vcf.freebayes.HQ.1.Ae <- vcf.freebayes.HQ.Ae[(!is.na(vcf.freebayes.HQ.Ae$Ae_tot.depth) &
vcf.freebayes.HQ.Ae$Ae_tot.depth > 10 &
vcf.freebayes.HQ.Ae$Ae_tot.depth < 1000),]
tmp.list <- strsplit(vcf.freebayes.HQ.1.Ae$INFO,split = ";")
for (i in 1:length(tmp.list)){
tmp.list[[i]] <- tmp.list[[i]][c(1:41)]
}
info <- matrix(
unlist(tmp.list),
nrow=nrow(vcf.freebayes.HQ.1.Ae),
byrow=TRUE
)
colnames(info) <- c("AB","ABP","AC","AF","AN","AO","CIGAR","DP","DPB","DPRA","EPP","EPPR","GTI","LEN","MEANALT",
"MQM","MQMR","NS","NUMALT","ODDS","PAIRED","PAIREDR","PAO","PQA","PQR","PRO","QA","QR","RO",
"RPL","RPP","RPPR","RPR","RUN","SAF","SAP","SAR","SRF","SRP","SRR","TYPE")
for (i in 1:ncol(info)){
info[,i] <- gsub("([[:print:]]+)(=)([[:print:]])","\\3",info[,i])
}
vcf.freebayes.HQ.1.Ae <- cbind(vcf.freebayes.HQ.1.Ae,info,stringsAsFactors=FALSE)
# tail bias
vcf.freebayes.HQ.1.reform.Ae <- vcf.freebayes.HQ.1.Ae
vcf.freebayes.HQ.1.reform.Ae$RPR <- as.numeric(vcf.freebayes.HQ.1.reform.Ae$RPR)
vcf.freebayes.HQ.1.reform.Ae$RPL <- as.numeric(vcf.freebayes.HQ.1.reform.Ae$RPL)
vcf.freebayes.HQ.1.reform.Ae$MBR <- round(pmin(vcf.freebayes.HQ.1.reform.Ae$RPR, vcf.freebayes.HQ.1.reform.Ae$RPL)/(vcf.freebayes.HQ.1.reform.Ae$RPR + vcf.freebayes.HQ.1.reform.Ae$RPL), digits = 2) vcf.freebayes.HQ.2.tmp.Ae <- vcf.freebayes.HQ.1.reform.Ae
vcf.freebayes.HQ.2.test1.Ae <-
vcf.freebayes.HQ.2.tmp.Ae[(vcf.freebayes.HQ.2.tmp.Ae$MBR>0.01),]
# do the same analysis for Da-Ol-1 data, output file is vcf.freebayes.HQ.2.test1.Ol
# combine Da-Ae and Da-Ol-1 filtered vcf data
load("~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/data/freebayes/vcf.freebayes.HQ.2.test1.Ae")
load("~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/data/freebayes/vcf.freebayes.HQ.2.test1.Ol")
Ae <- vcf.freebayes.HQ.2.test1.Ae[,c("CHROM", "POS", "REF", "ALT", "Ae_gt")]
Ol <- vcf.freebayes.HQ.2.test1.Ol[,c("CHROM", "POS", "REF", "ALT", "Ol_gt")]
Ae$feature <- paste(Ae$CHROM, Ae$POS, Ae$REF, Ae$ALT, sep = "_")
Ol$feature <- paste(Ol$CHROM, Ol$POS, Ol$REF, Ol$ALT, sep = "_")
freebayes_result <- merge(Ae, Ol, by="feature", all=T)
freebayes_result_reform <- freebayes_result[,c("feature", "Ae_gt", "Ol_gt")]
freebayes_result_reform$Ae_gt <- gsub("NA", "1/1", freebayes_result_reform$Ae_gt)
freebayes_result_reform[is.na(freebayes_result_reform$Ae_gt)] <- "0/0"
freebayes_result_reform[which(is.na(freebayes_result_reform$Ae_gt)),]$Ae_gt <- "0/0"
freebayes_result_reform[which(is.na(freebayes_result_reform$Ol_gt)),]$Ol_gt <- "0/0"
freebayes_result_1 <- freebayes_result_reform[freebayes_result_reform$Ae_gt!=freebayes_result_reform$Ol_gt & freebayes_result_reform$Ae_gt!="0/1" & freebayes_result_reform$Ol_gt!="0/1",]
```
### SNP filtering for GATK output
```{r}
SNP.GATK.basic.filter <- function(vcf){
# filter based on snpcluster
vcf.pass <- vcf[vcf$FILTER!="SnpCluster",]
snpcluster.pass.ratio <- nrow(vcf.pass)/nrow(vcf)
# filter based on QUAL score
vcf.HQ <- vcf.pass[vcf.pass$QUAL>40,]
QUAL.40.pass.ratio <- nrow(vcf.HQ) / nrow(vcf.pass)
Ae.gt.matrix <- table(vcf.HQ$Ae_gt)
Ol.gt.matrix <- table(vcf.HQ$Ol_gt)
cat("the percentage of SNPs that are not in snpcluster:", snpcluster.pass.ratio, "\n")
cat("The percentage of SNPs with QUAL > 40:", QUAL.40.pass.ratio, "\n")
cat("genotyping call matrix for Ae:", "\n")
print(Ae.gt.matrix)
cat("genotyping call matrix for Ol:", "\n")
print(Ol.gt.matrix)
return(vcf.HQ)
}
vcf.GATK <- read.table("~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/data/GATK/as_diploid/no_bolting/GATK_SNP_biallelic.recode.ann.vcf",as.is=T,na.strings = ".")
vcf.header.GATK <- system("grep '#C' ~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/data/GATK/as_diploid/no_bolting/GATK_SNP_biallelic.recode.ann.vcf",intern = TRUE)
vcf.header.GATK <- sub("#","",vcf.header.GATK)
vcf.header.GATK <- unlist(strsplit(vcf.header.GATK,split="\t"))
vcf.GATK.reform <- SNP.GATK.reformat(vcf.GATK, vcf.header.GATK)
vcf.GATK.HQ <- SNP.GATK.basic.filter(vcf.GATK.reform)
# depth
vcf.GATK.HQ.1 <- vcf.GATK.HQ[(!is.na(vcf.GATK.HQ$Ae_approx.depth) &
!is.na(vcf.GATK.HQ$Ol_approx.depth) &
vcf.GATK.HQ$Ae_approx.depth > 10 &
vcf.GATK.HQ$Ol_approx.depth > 10 &
vcf.GATK.HQ$Ae_approx.depth < 1000 &
vcf.GATK.HQ$Ol_approx.depth < 1000),]
# genotype quality
vcf.GATK.HQ.2 <- vcf.GATK.HQ.1[(!is.na(vcf.GATK.HQ.1$Ae_genotype.qual) &
!is.na(vcf.GATK.HQ.1$Ae_genotype.qual) &
vcf.GATK.HQ.1$Ae_genotype.qual > 30 &
vcf.GATK.HQ.1$Ol_genotype.qual > 30),]
vcf.tmp <- vcf.GATK.HQ.2
# subset SNPs between Ae & Ol
vcf.Ae.Ol.GATK <- vcf.tmp[((vcf.tmp$Ae_gt=="1/1" & vcf.tmp$Ol_gt=="0/0") | (vcf.tmp$Ae_gt=="0/0" & vcf.tmp$Ol_gt=="1/1")),]
```
### get intersection between freebayes and GATK output
```{r}
load("~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/data/freebayes/freebayes_result_1.Rdata")
load("~/Desktop/Brassica_project/KIAT_RNA_seq/parent_SNP/output/vcf.Ae.Ol.GATK.Rdata")
freebayes_result_1$CHROM <- gsub("([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]])", "\\1", freebayes_result_1$feature)
freebayes_result_1$POS <- gsub("([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]])", "\\3", freebayes_result_1$feature)
freebayes_result_1$REF <- gsub("([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]])", "\\5", freebayes_result_1$feature)
freebayes_result_1$ALT <- gsub("([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]]+)(_)([[:print:]])", "\\7", freebayes_result_1$feature)
vcf.Ae.Ol.freebayes.tmp <- freebayes_result_1
vcf.Ae.Ol.GATK.tmp <- vcf.Ae.Ol.GATK
# goal: get intersections where there SNP calls for both Da-Ae & Da-Ol-1 are the same.
vcf.freebayes <- paste(vcf.Ae.Ol.freebayes.tmp$CHROM, vcf.Ae.Ol.freebayes.tmp$POS,
vcf.Ae.Ol.freebayes.tmp$REF, vcf.Ae.Ol.freebayes.tmp$ALT,
vcf.Ae.Ol.freebayes.tmp$Ae_gt, vcf.Ae.Ol.freebayes.tmp$Ol_gt,
sep = "_")
vcf.GATK <- paste(vcf.Ae.Ol.GATK.tmp$CHROM, vcf.Ae.Ol.GATK.tmp$POS,
vcf.Ae.Ol.GATK.tmp$REF, vcf.Ae.Ol.GATK.tmp$ALT,
vcf.Ae.Ol.GATK.tmp$Ae_gt, vcf.Ae.Ol.GATK.tmp$Ol_gt,
sep = "_")
vcf.Ae.Ol.intersect.2 <- intersect(vcf.freebayes, vcf.GATK)
length(vcf.Ae.Ol.intersect.2) # 63981 same SNP positions between freebayes & GATK
```
### SNP annotation using snpEff
```{r}
# 1) build database for B.napus
# download snpEff databse, unzip
# mkdir data
# cd data
# mkdir napus
# cd napus
# use the gff3 file in
# /Network/Servers/avalanche.plb.ucdavis.edu/Volumes/Mammoth/Users/ruijuanli/Reference/B.napus/Brassica_napus.annotation_v5_modified_modified.gff3 (modified gff3 file)
# mv Brassica_napus.annotation_v5_modified_modified.gff3 genes.gff
# wget ftp://bradata:zhl410ivf@brassicadb.org/Brassica_napus/Brassica_napus_v4.1.chromosomes.fa.gz
# mv Brassica_napus_v4.1.chromosomes.fa.gz sequences.fa.gz
# gunzip *.gz
# cd ..
# mkdir genomes
# cd genomes/
# wget ftp://bradata:zhl410ivf@brassicadb.org/Brassica_napus/Brassica_napus_v4.1.chromosomes.fa.gz
# mv Brassica_napus_v4.1.chromosomes.fa.gz sequences.fa.gz
# gunzip sequences.fa.gz
# cd .. (to snpEff home directory)
# vim snpEff.config ####
# add "# Brassica napus genome napus.genome: Brassica_napus"
# java -jar snpEff.jar build -gff3 -v napus
# 2) SNP annotation
# cd cd ~/bin/snpEff/
# java -jar snpEff.jar napus SNP_biallelic.recode.vcf > SNP_biallelic.recode.ann.vcf
```
### circos plot
```{r}
library(circlize)
setEPS()
postscript(file = "../output/density.eps", width=5.7, height=5.7)
par(mai = c(0.4, 0.4, 0.4, 0.4), ps = 8, cex = 1, cex.main = 1, cex.axis = 0.8, tcl = -0.3)
# add SNP data for the marker, cytoband format
F2_map <- read.csv("../input/LG.f2.madmapper.final.flipped_gen_C05C08.csv_gen.csv")
F2_map$chrom <- F2_map$X
F2_map$start <- gsub("(chr)(A|C)(01|02|03|04|05|06|07|08|09|10)(_)([[:print:]]+)", "\\5", F2_map$id) %>% as.numeric()
F2_map_cytoband <- data.frame(V1 = F2_map$chrom,
V2 = F2_map$start,
V3 = F2_map$start + 100000,
V4 = F2_map$id,
V5 = c("stalk"))
circos.initializeWithIdeogram(F2_map_cytoband)
# data
# add SNP data
vcf.Ae.Ol.intersect.df.2.sorted <- read.csv("../input/vcf.Ae.Ol.intersect.df.2.csv")
vcf.Ae.Ol.intersect.df.2.sorted$CHROM <- gsub("chr","",vcf.Ae.Ol.intersect.df.2.sorted$CHROM)
dim(vcf.Ae.Ol.intersect.df.2.sorted) # 62958 9
circos.trackHist(factors = vcf.Ae.Ol.intersect.df.2.sorted$CHROM, x = vcf.Ae.Ol.intersect.df.2.sorted$POS, bin.size = 1000000, col = "blue", border = NA, bg.col = "#EFEFEF")
gff.mRNA <- read.table("../input/gff.mRNA")
colnames(gff.mRNA) <- c("chrom", "start", "end", "gene_ID")
gff.mRNA$subgenome <- gsub("(chr)(A|C)(01|02|03|04|05|06|07|08|09|10)", "\\2", gff.mRNA$chrom)
gff.mRNA.main <- gff.mRNA[grep("random", gff.mRNA$chrom, invert = T),]
gff.mRNA.main$chrom <- gsub("(chr)(A|C)(01|02|03|04|05|06|07|08|09|10)", "\\3", gff.mRNA.main$chrom)
dim(gff.mRNA.main)
gff.mRNA.main$chrom <- paste(gff.mRNA.main$subgenome, gff.mRNA.main$chrom, sep = "")
circos.trackHist(factors = gff.mRNA.main$chrom, x = gff.mRNA.main$start, bin.size = 1000000,
col = "orange", border = NA, bg.col = "#EFEFEF")
dev.off()
```