Skip to content

README.md

Latest commit

 

History

History
89 lines (62 loc) · 2.57 KB

README.md

File metadata and controls

89 lines (62 loc) · 2.57 KB

dcHiChIP

dcHiChIP is a modular and scalable pipeline for the multi-scale analysis of chromatin architecture from HiChIP data, developed using Nextflow DSL2. It supports different experimental scenarios, handles a wide variety of input formats, and integrates state-of-the-art tools for loop calling, stripe detection, compartment analysis, and 3D genome modelling.

Key Features

The pipeline includes the following modules:

  • Read Alignment & Filtering (BWA, SAMtools, pairtools)
  • Peak Calling (MACS3)
  • Loop Calling (MAPS)
  • Stripe Calling (gStripe)
  • Compartment & TAD Analysis (cooltools)
  • Motif & Peak Annotation (HOMER, Bioframe)
  • Coverage & Correlation Plots (DeepTools)
  • 3D Genome Modelling (MultiMM)
  • Subcompartment Calling (CALDER)

Documentation

A complete and interactive documentation is available at: dcHiChIP Documentation

Example Use Cases

dcHiChIP supports three types of input design:

  • Case 1: HiChIP-only input → pseudo-ChIP signal
  • Case 2: HiChIP + peak files (.narrowPeak) from ChIP-seq
  • Case 3: HiChIP + raw ChIP-seq (with inputs) → calls peaks internally

Example datasets and sample designs are provided in the "Working Cases" section of documentation.

Input Requirements

To run dcHiChIP, you will need:

  • Paired-end HiChIP FASTQ files
  • Optionally:
    • ChIP-seq FASTQ files (with/without inputs), or
    • Pre-processed .narrowPeak files
  • A design CSV describing samples
  • BWA index for the reference genome
  • Chromosome sizes file
  • Genomic features file (e.g., BED gene annotation)

Quick Start

  1. Install Nextflow (>=22.10.1)

    bash
curl -s https://get.nextflow.io | bash

  2. Set up a software environment using one of:

  • Docker
  • Singularity (recommended for HPC)
  • Podman
  • Shifter
  • Charliecloud
  • Conda (only as a last resort)

Note - You can chain multiple config profiles, e.g., profile test, docker.

  1. Run your own analysis:
nextflow run SFGLab/dcHiChIP \
             -profile <docker/singularity/podman/shifter/charliecloud/conda/institute> \
             --input samplesheet.csv \
             --outdir results

Note -

  • Use nf-core/configs to check if your institute already has a config file.
  • For Singularity, pre-download images using nf-core download and set a cache directory (NXF_SINGULARITY_CACHEDIR).
  • For Conda, set a cache directory (NXF_CONDA_CACHEDIR) to avoid re-installing environments.