soccin / ChIP-seq

ChIP-seq pipeline

ChIP-Seq pipeline

Version 0.7.2

Differential Peak Reference

https://genomebiology.biomedcentral.com/articles/10.1186/s13059-022-02686-y

• Example run command:

bsub -o LSF.CTRL/ ./ChIP-seq/pipe.sh \
    --pairing-file results/Proj_10706_B_sample_pairing.txt \
    results/alignments/Proj_10706_B_s_*bam

The pairing file should be

Col-1	Col-2
CTRL_1	TRGT_1
CTRL_2	TRGT_2

You need to use a pairing file. If the samples are TARGET only (no control) then use 'na' for control:

Col-1	Col-2
na	TRGT_1
na	TRGT_2

Version which uses both reads from PE-runs. Using methods from R.K. for bigWig generation. Added option to allow the use of non-proper paired reads for cases where translocations important.

Try to match ENCODE 3 (https://github.com/soccin/ChIP-seq) as closely as possible

• Post-alignment filtering: Get from ENCODE 3

  • Code for filter_qc.py
  • samtools view -q 30 -f 2 -F 1804 -f 2
    • MAPQ<30
    • Proper Pair (can be turned off)
    • exclude: unmapped, mate unmapped, secondary align, failing QC, Duplicate
  • Also keep only the main chromosomes [1-(19|22),X,Y]; exclude MT, and the rest of the build, decoys, viral, ...

• BigWig file generation.

  • convert bam to bed
  • compute density for bigwig formation
    • normalizing to 10 million mapped reads