Chromap output 理解

• Understanding The Chromap Summary File
  

  -

• （Chromap - 简书）preset 模式到atac，基本的处理过程是trim3'端的接头，比对，细胞水平去重、做ATAC的 peak shift，然后根据提供的barcode 白名单进行barcode矫正。

Chromap will perform all of the following steps automatically for us:

• Trim adapter sequences at the 3’ end of the reads
• Set the maximum insert size between Read 1 & Read 2 to be 2000 bp
• Deduplicate the mapped read pairs at the single-cell level
• Correct cell barcodes based on provided whitelist
• Shift read positions to make the start positions (5’) of the reads represent the actual Tn5 binding sites (not important for downstream peak calling, though)
• Output read pairs (MAPQ >= 30 by default) as the commonly-used BED-like file fragments.tsv. In addition, the SAM format is also supported.

chromap -t 20 -x ref.idx -r ref.fa \
        --preset atac --summary summary.csv \
        -1 mCortex_ATAC_S1_L001_R1_001.fastq.gz \
        -2 mCortex_ATAC_S1_L001_R2_001.fastq.gz \
        -b mCortex_ATAC_S1_L001_I2_001.fastq.gz \
        --barcode-whitelist 737K-cratac-v1_rc.txt \
        -o chromap_outs/fragments.tsv \
        1> chromap.stdout 2> chromap.stderr \
        && bgzip chromap_outs/fragments.tsv

两部分结果

1. chromap.stderr
2. summary.csv

image.png

• Summary 文件行数：从barcode fastq 中识别到的所有barcode
• Raw reads 数 ：475,581,592 （*2= 951,163,184）
• Summary-total：每个barcode序列的reads数（reads/cell）
• Number of mapped reads：810,898,762 = 792,439,756 + 18,459,006 ； is consistent with the summary.csv file, which is (total - unmapped) * 2:(475,581,592 - 70,132,211) × 2 = 810,898,762
• Number of output mappings (passed filters): 161219326 (fragment.tsv 行数)
• 输出的细胞的fragment条数=total - unmapped - lowmapq - duplicate

Overall alignment (mapping) rate:(total - unmapped)/total
Total unique fragments:total - duplicate - unmapped - lowmapq
Duplication rate (saturation):duplicate/(total - unmapped - lowmapq)