這里是佳奧！新的一年，ATAC-Seq的學(xué)習(xí)也進(jìn)入了尾聲。
讓我們開(kāi)始吧！
1 peaks注釋
統(tǒng)計(jì)peak在promoter，exon，intron和intergenic區(qū)域的分布
QQ截圖20230101155625.png
if(F){
options(BioC_mirror="https://mirrors.ustc.edu.cn/bioc/")
options("repos" = c(CRAN="https://mirrors.tuna.tsinghua.edu.cn/CRAN/"))
source("http://bioconductor.org/biocLite.R")
BiocManager::install('TxDb.Mmusculus.UCSC.mm10.knownGene')
BiocManager::install('org.Mm.eg.db')
}
bedPeaksFile = '2-ceLL-1_peaks.narrowPeak';
bedPeaksFile
## loading packages
require(ChIPseeker)
require(TxDb.Mmusculus.UCSC.mm10.knownGene)
txdb <- TxDb.Mmusculus.UCSC.mm10.knownGene
require(clusterProfiler)
peak <- readPeakFile( bedPeaksFile )
##去除含_的染色體
keepChr= !grepl('_',seqlevels(peak))
seqlevels(peak, pruning.mode="coarse") <- seqlevels(peak)[keepChr]
peakAnno <- annotatePeak(peak, tssRegion=c(-3000, 3000),
TxDb=txdb, annoDb="org.Mm.eg.db")
peakAnno_df <- as.data.frame(peakAnno)
promoter <- getPromoters(TxDb=txdb, upstream=3000, downstream=3000)
tagMatrix <- getTagMatrix(peak, windows=promoter)
# 然后查看這些peaks在所有基因的啟動(dòng)子附近的分布情況，熱圖模式
tagHeatmap(tagMatrix, xlim=c(-3000, 3000), color="red")
# 然后查看這些peaks在所有基因的啟動(dòng)子附近的分布情況，信號(hào)強(qiáng)度曲線圖
plotAvgProf(tagMatrix, xlim=c(-3000, 3000),
xlab="Genomic Region (5'->3')", ylab = "Read Count Frequency")
plotAnnoPie(peakAnno)
QQ截圖20230101160301.png
QQ截圖20230101160325.png
QQ截圖20230101160332.png
可以載入IGV看看效果，檢測(cè)軟件找到的peaks是否真的合理，還可以配合rmarkdown來(lái)出自動(dòng)化報(bào)告。
https://ke.qq.com/course/274681
我們可以看到Tcea1基因轉(zhuǎn)錄起始位置有peaks富集
QQ截圖20230101160746.png
也可以使用其它代碼進(jìn)行下游分析；
https://github.com/jmzeng1314/NGS-pipeline/tree/master/CHIPseq
Homer 可以做，但是需要下載數(shù)據(jù)庫(kù)
# perl ~/miniconda3/envs/atac/share/homer-4.9.1-5/configureHomer.pl -install mm10
# ln -s /home/jmzeng/miniconda3/envs/chipseq/share/homer-4.9.1-5/data/genomes/ genomes
# cp /home/jmzeng/miniconda3/envs/chipseq/share/homer-4.9.1-5/config.txt /home/stu/miniconda3/envs/atac/share/homer-4.9.1-5/config.txt
## 保證數(shù)據(jù)庫(kù)下載是OK
ls -lh ~/miniconda3/envs/atac/share/homer-4.9.1-5/data/genomes
mkdir -p ~/project/atac/peaks
source activate atac
cd ~/project/atac/peaks
ls *.narrowPeak |while read id;
do
echo $id
awk '{print $4"\t"$1"\t"$2"\t"$3"\t+"}' $id >{id%%.*}.homer_peaks.tmp
annotatePeaks.pl {id%%.*}.homer_peaks.tmp mm10 1>${id%%.*}.peakAnn.xls
2>${id%%.*}.annLog.txt
done
Bedtools也可以做
https://bedtools.readthedocs.io/en/latest/content/tools/annotate.html
2 motif尋找及注釋
Homer可以做
ls -lh ~/miniconda3/envs/atac/share/homer-4.9.1-5/data/genomes
mkdir -p ~/project/atac/motif
cd ~/project/atac/motif
source activate atac
ls ../peaks/*.narrowPeak |while read id;
do
file=$(basename $id )
sample=${file%%.*}
echo $sample
awk '{print $4"\t"$1"\t"$2"\t"$3"\t+"}' $id > ${sample}.homer_peaks.tmp
nohup findMotifsGenome.pl ${sample}.homer_peaks.tmp mm10 ${sample}_motifDir -len 8,10,12 &
done
meme 也可以做 ，首先利用.bed獲取.fa序列：
https://github.com/jmzeng1314/NGS-pipeline/blob/master/CHIPseq/step7-peaks2sequence.R
##usage: Rscript peakView.R peaks.bed IP.sorted.bam input.sorted.bam 10
#options(echo=TRUE) # if you want see commands in output file
args <- commandArgs(trailingOnly = TRUE)
if(length(args) != 1 ){
print(" usage: Rscript peakAnno.R peaks.bed ")
}
bedPeaksFile = args[1] ;
##自這開(kāi)始，.bed文件要和R Project文件在同一目錄下
bedFiles=list.files(pattern = '*.bed')
> bedFiles
[1] "2-ce11-2_summits.bed" "2-ce11-4_summits.bed" "2-ce11-5_summits.bed" "2-ceLL-1_summits.bed"
BiocManager::install("BSgenome.Mmusculus.UCSC.mm10")
library(BSgenome.Mmusculus.UCSC.mm10)
library(ChIPpeakAnno)
##生成.fa文件
bedPeaksFile=bedFiles[2]##第二個(gè)文件即2-ce11-4_summits.bed，要下一個(gè)就[3]
sampleName=strsplit(bedPeaksFile,'\\.')[[1]][1]
peak <- toGRanges(bedPeaksFile, format="BED")
keepChr= !grepl('_',seqlevels(peak))
#seqlevels(peak, force=TRUE) <- seqlevels(peak)[keepChr]
seq <- getAllPeakSequence(peak, upstream=20, downstream=20, genome=Mmusculus)
write2FASTA(seq, paste0(sampleName,'.fa'))
使用網(wǎng)頁(yè)端注釋
https://meme-suite.org/meme/
QQ截圖20230101170341.png
QQ截圖20230101170412.png
R包，比如
motifmatchr包 也可以做。
https://bioconductor.org/packages/release/bioc/html/motifmatchr.html
3 多組學(xué)整合分析
RNS-Seq、ChIP-Seq、ATAC-Seq
以及一些整合的R包：esATAC
ATAC-Seq的實(shí)戰(zhàn)流程學(xué)習(xí)至此結(jié)束。
但個(gè)性化的分析還有很多要鉆研的地方，尤其是官方文檔。
新年快樂(lè)！我們下一個(gè)篇章再見(jiàn)！
