deepTools

deepTools 是一套基于python開(kāi)發(fā)的工具，適用于有效處理分析高通量測(cè)序數(shù)據(jù)，可用于ChIP-seq, RNA-seq 或 MNase-seq。

#1. deepTools 系列工具

deepTools workflow

##1.1 deepTools 系列工具信息匯總

##1.2 BAM 和bigWig文件處理工具

• multiBamSummary

? 利用兩個(gè)或多個(gè)bam文件計(jì)算基因組區(qū)段reads覆蓋度；BED-file 指定基因組區(qū)域，bins模式可用于全基因組范圍分析；產(chǎn)生的結(jié)果(.npz)可用于plotCorrelation進(jìn)行相關(guān)性分析和用于plotPCA進(jìn)行主成分分析。

$ deepTools2.0/bin/multiBamSummary bins \
 --bamfiles testFiles/*bam \ # using all BAM files in the folder
 --minMappingQuality 30 \
 --region 19 \ # limiting the binning of the genome to chromosome 19
 --labels H3K27me3 H3K4me1 H3K4me3 HeK9me3 input \
 -out readCounts.npz --outRawCounts readCounts.tab

 $ head readCounts.tab
 #'chr'     'start' 'end'   'H3K27me3'      'H3K4me1'       'H3K4me3'       'HeK9me3'       'input'
 19 10000   20000   0.0     0.0     0.0     0.0     0.0
 19 20000   30000   0.0     0.0     0.0     0.0     0.0
 19 30000   40000   0.0     0.0     0.0     0.0     0.0
 19 40000   50000   0.0     0.0     0.0     0.0     0.0
 19 50000   60000   0.0     0.0     0.0     0.0     0.0
 19 60000   70000   1.0     1.0     0.0     0.0     1.0
 19 70000   80000   0.0     1.0     7.0     0.0     1.0
 19 80000   90000   15.0    0.0     0.0     6.0     4.0
 19 90000   100000  73.0    7.0     4.0     16.0    5.0

• multiBigwigSummary
  ? 與multiBamSummary相比，輸入文件格式是bigWig 。

• correctGCBias
  ? 矯正GC-bias；

• bamCoverage
  

  norm_IGVsnapshot_indFiles
  
  ? bamCoverage 利用測(cè)序數(shù)據(jù)比對(duì)結(jié)果轉(zhuǎn)換為基因組區(qū)域reads覆蓋度結(jié)果?？梢宰孕性O(shè)定覆蓋度計(jì)算的窗口大小(bin)；bamCoverage 內(nèi)置了各種標(biāo)準(zhǔn)化方法：scaling factor, Reads Per Kilobase per Million mapped reads (RPKM), counts per million (CPM), bins per million mapped reads (BPM) and 1x depth (reads per genome coverage, RPGC).

Example : bamCoverage 用于ChIPseq分析

bamCoverage --bam a.bam -o a.SeqDepthNorm.bw \
    --binSize 10
    --normalizeUsing RPGC
    --effectiveGenomeSize 2150570000
    --ignoreForNormalization chrX
    --extendReads
    --outFileFormat bedgraph

• bamCompare
  ? 兩個(gè)BAM 文件相比較，計(jì)算二者之間窗口中的reads豐度比率。

usage:  bamCompare -b1 treatment.bam -b2 control.bam -o log2ratio.bw

• bigwigCompare
• computeMatrix
  ? 給基因組區(qū)段打分，產(chǎn)生的文件可用于plotHeatmap 和plotProfiles作圖；基因組區(qū)段可以是基因或其他區(qū)域，使用BED格式文件定義即可。

computeMatrix 有兩種不同的模式

computeMatrix two modes

• reference-point(relative to a point): 計(jì)算某個(gè)點(diǎn)的信號(hào)豐度
• scale-regions(over a set of regions): 把所有基因組區(qū)段縮放至同樣大小，然后計(jì)算其信號(hào)豐度
  如下命令查看幫助：

$ computeMatrix scale-regions –help
$ computeMatrix scale-regions -S <biwig file(s)> -R <bed file> -b 1000
$ computeMatrix reference-point –help
$ computeMatrix reference-point -S <biwig file(s)> -R <bed file> -a 3000 -b 3000

Example 1：?jiǎn)蝹€(gè)輸入文件 (reference-point mode)

$ computeMatrix reference-point \ # choose the mode
       --referencePoint TSS \ # alternatives: TSS, TES, center
       -b 3000 -a 10000 \ # define the region you are interested in
       -R testFiles/genes.bed \
       -S testFiles/log2ratio_H3K4Me3_chr19.bw  \
       --skipZeros \
       -o matrix1_H3K4me3_l2r_TSS.gz \ # to be used with plotHeatmap and plotProfile
       --outFileSortedRegions regions1_H3K4me3_l2r_genes.bed

? 注：point-BED文件指定基因組區(qū)段的起始位置

Example 2：多個(gè)輸入文件 (scale-regions mode)

$ deepTools2.0/bin/computeMatrix scale-regions \
  -R genes_chr19_firstHalf.bed genes_chr19_secondHalf.bed \ # separate multiple files with spaces
  -S testFiles/log2ratio_*.bw  \ or use the wild card approach
  -b 3000 -a 3000 \
  --regionBodyLength 5000 \
  --skipZeros -o matrix2_multipleBW_l2r_twoGroups_scaled.gz \
  --outFileNameMatrix matrix2_multipleBW_l2r_twoGroups_scaled.tab \
  --outFileSortedRegions regions2_multipleBW_l2r_twoGroups_genes.bed
Note that the reported regions will have the same coordinates as the ones in 

##1.3 質(zhì)控工具

• plotCorrelation
  ? 基于multiBamSummary 或multiBigwigSummary結(jié)果計(jì)算樣品間的相關(guān)性。并且還可以通過(guò)Scatterplot或Heatmap進(jìn)行展示。

Example 1：Scatterplot

$ deepTools2.0/bin/plotCorrelation \
-in scores_per_transcript.npz \
--corMethod pearson --skipZeros \
--plotTitle "Pearson Correlation of Average Scores Per Transcript" \
--whatToPlot scatterplot \
-o scatterplot_PearsonCorr_bigwigScores.png   \
--outFileCorMatrix PearsonCorr_bigwigScores.tab

scatterplot_PearsonCorr_bigwigScores

Example 2：Heatmap

$ deepTools2.0/bin/plotCorrelation \
    -in readCounts.npz \
    --corMethod spearman --skipZeros \
    --plotTitle "Spearman Correlation of Read Counts" \
    --whatToPlot heatmap --colorMap RdYlBu --plotNumbers \
    -o heatmap_SpearmanCorr_readCounts.png   \
    --outFileCorMatrix SpearmanCorr_readCounts.tab

heatmap_SpearmanCorr_readCounts1

• plotPCA
  ? 基于multiBamSummary 或multiBigwigSummary結(jié)果進(jìn)行主成分分析，并作出基于兩個(gè)主成分的圖和前五個(gè)特征代表性的圖。

Example

$ deepTools2.0/bin/plotPCA -in readCounts.npz \
-o PCA_readCounts.png \
-T "PCA of read counts"

PCA_readCounts

• plotFingerprint
  ? 對(duì)樣本比對(duì)結(jié)果reads累積情況進(jìn)行展示。一定長(zhǎng)度窗口(bin)上reads數(shù)進(jìn)行計(jì)數(shù)，然后排序，再依次累加畫(huà)圖。input (能測(cè)到90%DNA片段)在基因組理論上是均勻分布，隨著測(cè)序深度增加趨近于直線，實(shí)驗(yàn)組在排序越高的窗口處reads累積速度越快，說(shuō)明這些區(qū)域富集的越特異。
  
  QC_fingerprint

Example

$ deepTools2.0/bin/plotFingerprint \
 -b testFiles/*bam \
--labels H3K27me3 H3K4me1 H3K4me3 H3K9me3 input \
--minMappingQuality 30 --skipZeros \
--region 19 --numberOfSamples 50000 \
-T "Fingerprints of different samples"  \
--plotFile fingerprints.png \
--outRawCounts fingerprints.tab

fingerprints1

• bam PEFragmentSize
  ? 計(jì)算bam文件中雙端reads的fragment size長(zhǎng)度。
• compute GCBias
  ? 計(jì)算GC-bias
• plot Coverage
  ? 計(jì)算樣品測(cè)序深度。隨機(jī)抽取1 million bp ，計(jì)算reads數(shù)，統(tǒng)計(jì)堿基覆蓋率和覆蓋次數(shù)。

##1.4 熱圖和總結(jié)圖

• plotHeatmap

Example 1: 根據(jù)computeMatrix結(jié)果畫(huà)熱圖

# run compute matrix to collect the data needed for plotting
$ computeMatrix scale-regions -S H3K27Me3-input.bigWig \
                                 H3K4Me1-Input.bigWig  \
                                 H3K4Me3-Input.bigWig \
                              -R genes19.bed genesX.bed \
                              --beforeRegionStartLength 3000 \
                              --regionBodyLength 5000 \
                              --afterRegionStartLength 3000
                              --skipZeros -o matrix.mat.gz
$ plotHeatmap -m matrix.mat.gz \
      -out ExampleHeatmap1.png \

plot Heatmap

Example 2: plotHeatmap還可以進(jìn)行聚類分析

$ plotHeatmap -m matrix_two_groups.gz \
     -out ExampleHeatmap2.png \
     --colorMap RdBu \
     --whatToShow 'heatmap and colorbar' \
     --zMin -3 --zMax 3 \
     --kmeans 4  #聚類參數(shù)

[圖片上傳失敗...(image-d28547-1556115525034)]
其他參數(shù)

顏色自定義：--colorList 'black, yellow' 'white,blue' '#ffffff,orange,#000000'
去掉熱圖邊框：--boxAroundHeatmaps no

• plotProfile
  根據(jù)computeMatrix結(jié)果畫(huà)圖。

Example 1: 根據(jù)樣本畫(huà)圖

# run compute matrix to collect the data needed for plotting
$ computeMatrix scale-regions -S H3K27Me3-input.bigWig \
                                 H3K4Me1-Input.bigWig  \
                                 H3K4Me3-Input.bigWig \
                              -R genes19.bed genesX.bed \
                              --beforeRegionStartLength 3000 \
                              --regionBodyLength 5000 \
                              --afterRegionStartLength 3000
                              --skipZeros -o matrix.mat.gz

$ plotProfile -m matrix.mat.gz \
              -out ExampleProfile1.png \
              --numPlotsPerRow 2 \
              --plotTitle "Test data profile"

Example Profile1

Example 2: 根據(jù)基因畫(huà)圖

$ plotProfile -m matrix.mat.gz \
     -out ExampleProfile2.png \
     --plotType=fill \ # add color between the x axis and the lines
     --perGroup \ # make one image per BED file instead of per bigWig file
     --colors red yellow blue \
     --plotTitle "Test data profile"

Example Profile2

Example 3: 聚類畫(huà)圖

$ plotProfile -m matrix.mat.gz \
      --perGroup \
      --kmeans 2 \
      -out ExampleProfile3.png

Example Profile3

Example 4: 畫(huà)熱圖

$ plotProfile -m matrix.mat.gz \
      --perGroup \
      --kmeans 2 \
      -plotType heatmap \
      -out ExampleProfile3.png

Example Profile4

Example

$ plotEnrichment -b Input.bam H3K4Me1.bam H3K4Me3.bam \
--BED up.bed down.bed \
--regionLabels "up regulated" "down regulated" \
-o enrichment.png

plot Enrichment

參考：

The tools of deeptools