-
install nextflow
curl -fsSL get.nextflow.io | bash
For details, visit https://www.nextflow.io/
-
clone this git repository
#for latest experimental features git clone https://github.com/CCBR/ChIP-Seq-Pipeline.git
For the stable version of the pipeline, download from the following link https://github.com/CCBR/ChIP-Seq-Pipeline/releases/latest
./nextflow run main.nf \
-config config \
--macsconfig='example/macs.config' \
--reads='example/*.fastq' \
--genome='hg19'
-
Note that the reads names are supposed to end with ".fastq" or ".fastq.gz". If you experience an error message related to missing file(s), this might be the reason.
-
Config file is designed to work at our NIH biowulf2 HPC. If you plan to use on helix, add -profile 'local' option at the end of command line arguments. If you plan to use outside NIH, you need to edit the "genome" section according to your paths information.
-
macs.config is a config file for MACS2 to find peaks. ChIP sample and control (input or IgG) with their label concatenated by comma form a line, e.g.
chip_sample_id,control_id,sample_name
.chip_sample_id
orcontrol_id
are the basename of the FASTQ files wihtout the trailing ".fastq" or ".fastq.gz". -
Currently, the pipeline fully supports hg19, mm10 and partially supports GRCh38 and GRCm38. Contact us for other genomes.
-resume
--> to resume the previous failed run-profile 'local'
--> for running tools "locally" not thourgh high performance computer queueing mechanisms.-with-timeline 'timeline.html'
--> record the run time of each process.-with-dag 'flowchart.png'
--> draw the flowchart--outdir 'ChIP-Seq-Pipeline-Output'
--> specify the directory to save directory from the default output
We implemented the pipeline using Nextflow.
Thanks to the authors of the tools!
- Trimgalor
- BWA mem
- Picard (MarkDuplicate)
- FASTQC
- DeepTools
- Macs2
- Sicer
- MEME-ChIP
- PhantomPeakqualTool
- CEAS
- ChipSeeker
- Homer
- PeakDiff
- ngsplot
I got many nice implementation ideas from Nextflow examples, especially from NGI-ChIP-seq pipeline. Many thanks to the NGI pipeline developers and Alexei, Ashley, George, Parthav and other CCBR team mebers.
Email to Bong-Hyun.Kim at NIH dot GOV.