Skip to main content

Capture-C Pipeline Overview

This section covers the analysis pipeline for Capture-C data, used to study long-range chromatin interactions at specific genomic regions of interest.

What is Capture-C?

Capture-C (also known as Next-Generation Capture-C or NG Capture-C) is a 3C-based method that captures chromatin interactions between a viewpoint (region of interest) and the rest of the genome. It’s particularly useful for studying enhancer-promoter interactions and regulatory elements.

Considerations for Capture-C Analysis

  • Restriction Enzyme: The choice of restriction enzyme used in the Capture-C experiment will affect the resolution of the interaction data and if you can view certain interactions depending on the configuration of the restriction fragments. Commonly used enzymes include DpnII, NlaIII, and HindIII.
  • Viewpoint Design: The design of the capture oligos is crucial for the success of the experiment.
    • Ensure that the oligos are specific to the region of interest, do not have off-target binding
    • Avoid repetitive sequences and regions with high GC content
    • Do not mix enhancers and promoters in the same capture oligo pool, as this can lead to competition for binding and reduce the efficiency of capture.
    • When trying to capture SNPs, create oligos that contain the ref and alt alleles, as this will allow you to capture both alleles and assess for allele-specific interactions.
    • Place the oligos as close to the restriction fragment ends as possible, as this will increase the chances of capturing interactions with the fragment ends.
    • Oligos should be ~120 bp in length and ideally there should be 2 per restriction fragment, one on each side of the fragment.
  • Sequencing Depth: The depth of sequencing required will depend on the complexity of the library and the number of viewpoints being captured. Generally, a higher depth of sequencing will provide more robust interaction data.
    • It is recommended to sequence at least 2 million reads per viewpoint per sample, but this may vary.
    • Per viewpoint, expect over 10,000 unique reporter fragments to be captured, which will allow you to generate robust interaction profiles.
    • Many reads will be PCR duplicates, which are removed during analysis.
  • Read Length: The read length should be sufficient to cover the captured fragments and allow for accurate mapping to the reference genome, as the sequence around the restriction sites guide the mapping and reporting of interactions. Paired-end reads are typically used to improve mapping accuracy and capture longer fragments.

For more information on Capture-C, see the original paper by Davies et al. (2016) Multiplexed analysis of chromosome conformation at vastly improved sensitivity.

And the method paper by Downes et al. (2022) Capture-C: a modular and flexible approach for high-resolution chromosome conformation capture.

Indeed this method has already been superseded by MCC (Micro-Capture-C) which is a more sensitive method that allows for the capture of interactions at a higher resolution, and can be used to study interactions at the level of individual nucleosomes. So consider which method is best for your research question and the resolution you require.

Pipeline Overview

Preparation steps:

  1. Prepping Fastq - Fastq contatenation (if multi-lane)
  2. Restriction fragment map - Generation of a restriction fragment map for the reference genome based on the enzyme used for digestion (genome_digest.pl)

Workflow steps:

  1. Quality Control - FastQC analysis of raw reads
  2. Read Processing - Adapter removal and FLASHing (Trimgalore and FLASH)
  3. Concatenation of trimmed merged and unmerged reads (keeping read order intact)
  4. In silico digestion - of reads based on the restriction enzyme used for CaptureC (fastq_digest.pl)
  5. Alignment - alignment of read fragments to a reference genome with Bowtie (with -p 1 -m 2 –best –strata –chunkmbs 256 to make sure reads are kept in the same order as the input fastq)
  6. Restriction Fragment Assignment - (CCanalyser.pl)
    • Labels a read a capture if it maps to a restriction fragment that contains a viewpoint oligo
    • Labels the other part of the read as a reporter if it maps to a restriction fragment that does not contain a viewpoint oligo
    • Labels read as proximity excluded if it maps to a restriction fragment that is within 1 kb of the viewpoint oligo
    • Also labels PCR duplicates based on the restriction fragment coordinates and the read sequence
  7. CaptureC Checks - Collates CaptureC reports
  8. Concatenates all .gffs from each viewpoint per sample
  9. Visualization
    • Generates BigWigs for each viewpoint per sample (raw and normalised to per 100,000 reads in cis)
    • Generates smoothed, binneda and normalised interaction profiles for each viewpoint averaged across samples per condition.

Outputs

  • QC reports
  • Interaction matrices
  • Visualization tracks
  • DESeq2 results for differential interactions between conditions