chip-seq-pipeline

This is the ChIP-seq mapping pipeline for our lab. The script is still at the very preliminary stage, please let us know how to improve!

How to run it.

Single sample.

1. Install snakemake
2. Go to your work directory, create and copy/link your fastq files in the fastq folder
3. /run_chipseq_vanilla.sh -g hg19 -e [[email protected]]. It will process every sample fastqs in the fastq/ directory, and send an e-mail when it is done.

Output files:

All the output files will be relative to your project directory.

1. bam/[name].filt.nodup.srt.bam: the final bam files after filter, duplicate removal, and sort.
2. bigWig/[name].filt.nodup.srt.bw: bigWig file for the bam.
3. qc: summary statistics for the bam files.
   1. *.flagstat.qc: summary stats for different steps of bam files.
   2. [name].*.cc.qc: phantom peak summary.
   3. [name].*.pbc.qc: PBC summary.
   4. [name].*.plot.pdf: tag shift size estimate plot.
4. logs/: running logs for different parts of the pipeline

Dependencies

• BWA
• samtools
• bedtools
• picard
• phantompeaktools

Plans for implementation

At its complete form, this pipeline will include work flow for analysis of (1) a single sample, and (2) replicates using the ENCODE IDR method.

Roadmap

2017.05.26: The Snakemake pipeline for single sample is available. 2017.05.17: Currently we have only the workflow for a single sample. And it is not modularized yet.