Skip to content

Latest commit

 

History

History
67 lines (42 loc) · 4.24 KB

README.md

File metadata and controls

67 lines (42 loc) · 4.24 KB

.

We have a new bioinformatic resource that largely replaces the functionality of this project! See our new repository here: https://github.com/epi2me-labs/pore-c-py.

We recommend that you use the tool through the associated nextflow pipeline: https://github.com/epi2me-labs/wf-pore-c.

This repository is now unsupported and we do not recommend its use. Please contact Oxford Nanopore: [email protected] for help with your application if it is not possible to upgrade to our new resources, or we are missing key features.

Pore-C tools

This package is designed to analyse the data from multi-contact pore-C reads. It is similar to the pairtools package in scope, however it is specifically designed to handle multi-contact reads (aka c-walks). It carries out the following operations:

  • pre-processing a reference genome to generate auxiliary files used in downstream analyses
  • creating virtual digests of the reference genome
  • processing BAM files to filter spurious alignments, detect ligation junctions and assign fragments
  • converting the resulting contacts to a pairs format and a COO-formatted matrix compatible with Cooler for downstream processing.

The recommended way to run Pore-C tools is through the associated Pore-C-Snakemake pipeline. It wraps the pore-C commands and also handles some of the analysis steps outside the scope of pore-C tools such as read alignment and conversion of output files to .cool format. If you want to run the tools outside of this pipeline please see the section below on

Sample data

There are some sample datasets (fastq, alignment parquets, .pairs, .cool files) available for HindIII-digested HG002 (31Gb) here and for NlaIII-digested GM12878 (23Gb) here.

Development

The recommended way to use pore-C tools is through the associated Pore-C-Snakemake pipeline, however if you want to run tools outside of this context you'll need to set up a development environment. We use the tox automation tool with the tox-conda extension to create conda environments for development. Several environments are defined in the tox.ini file:

  • py37: The default environment. Running tox -e py37 will run all of the pytest tests. By adding -- to the command you you can pass additional arguments to pytest e.g. tox -e py37 -- -k refgenome.
  • dev: A development environment that runs the pore_c command. You can use this to try our pore_c commands e.g. tox -e dev -- refgenome --help.
  • snakemake: An environment that runs an associated snakemake pipeline in development mode. The command assumes that the pore-c-snakemake repository is located in the same directory as pore-c though this can be changed with enviroment variables - see the tox.ini file for details.

For each of these environments a conda environment is created at .tox/<env_id>. You can activate these for interactive work with conda activate .tox/<env_id>.

Developing the Snakemake pipeline

If you're working on the Pore-C-Snakemake pipeline:

git clone [email protected]:nanoporetech/pore-c.git
cd pore-c
tox -e snakemake --notest

cd ../
git clone [email protected]:nanoporetech/Pore-C-Snakemake.git
cd pore-c-snakemake
conda activate ../pore-c/.tox/snakemake
snakemake -j 8 test --use-conda --config pore_c_version=dev

Citation

The biorxiv pre-print describing Pore-C can be found here:

Nanopore sequencing of DNA concatemers reveals higher-order features of chromatin structure