GangCaoLab / DLO-HiC-Tools

Tools for DLO HiC data analyze

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

DLO-HiC-Tools

Integrated DLO HiC analyze pipeline.

Installation

This package is only support UNIX-like system.

Recommend install and manage requirements with conda and mamba, just use following commands:

$ git clone https://github.com/Nanguage/DLO-HiC-Tools.git
$ cd DLO-HiC-Tools/
$ conda install mamba -c conda-forge  # skip this line if mamba is installed
$ mamba env create -f environment.yml
$ conda activate dlo-hic-tools
$ python setup.py install

Using Docker

You can also using the docker, pull the dlohic image:

$ docker pull nanguage/dlohic

Run a container, with mount current directory in file system to the '/data' in the container:

$ docker run -v $PWD:/data -ti nanguage/dlohic:latest

Quick Start

DLO-HiC-Tools is an integrated automatic pipeline for DLO-HiC data analyze. Before we start, you should install package and some dependency. see installation.

DLO-HiC-Tools implemented some tools for DLO-HiC data analyze and quality control. You can list them and get more usage detail by dlohic command:

$ dlohic --help
Usage: dlohic [OPTIONS] COMMAND [ARGS]...

  DLO HiC Tools command line interface.

Options:
  --log-file TEXT       The log file, default output to stdout.
  --debug / --no-debug  If debug will print all information, default False.
  --version             Show the version and exit.
  --help                Show this message and exit.

Commands:
  PET_span_dist         Count the distribution of PET span.
  bedpe2pairs           Transform BEDPE format file to pairs format,...
  build_bedpe           Build bedpe file from fastq files.
  chr_interactions      Count the interactions between chromosomes.
  extract_PET           Extract the PETs sequences on both sides of...
  extract_fragments     Extract all DNA fragments from fasta file,...
  fragment_length_dist  Draw the distribution figure(kde/box plot),...
  gen_qc_report         Generate pipeline quality control report.
  infer_adapter         Inference adapter sequence, and plot the...
  interactions_qc       Count ratio of: inter-chromosome...
  noise_reduce          Remove noise in DLO-HiC data.
  pipeline              Generate integrated main processes...
  plot_contact_map      Draw the Hi-C contact map.
  remove_redundancy     Remove the redundancy within pairs.

Use the pipeline

DLO-HiC-Tools implemented the pipeline with Snakemake. If you want generate the HiC matrix and the quality control report(see example) at once, you can use this pipeline.

The pipeline contain two files, a Snakemake file and a configuration file.

Firstly, generate them by dlohic pipeline command. It will generate the necessary files under your working directory.

$ dlohic pipeline
dlo_hic.tools.helper.pipeline INFO    @ 08/21/18 21:40:30: Generate config file at ./pipeline_config.ini
dlo_hic.tools.helper.pipeline INFO    @ 08/21/18 21:40:30: Generate pipeline (Snakemake file) at ./Snakefile
$ ls
pipeline_config.ini  Snakefile

Then edit the pipeline_config.ini file, just fill the sections with it's comment information. And you can reference the example data and the configuration file, see here.

After all necessary information wrote in to the configuration file, You can run the pipeline.

$ snakemake -j 16

The -j parameter indicate the number of jobs, it depend on your cpu cores number. Or, if you use the cluster like pbs system, you can run like this:

$ snakemake --cluster qsub -j 16

In addition, you can visualize the pipeline(need Graphviz installed):

$ snakemake all --dag | dot -Tpng > pipeline.png

doc/img/pipeline.png

More information about the Snakemake, please read it's document.

Visualize the contact map:

The pipeline can generate two kind of result matrix format, .hic and .cool. Although there are some contact maps in the HTML report, but if you want to browse the contact at finer resolution. You should use the software like:

  • Juicebox provided the GUI for .hic file.
  • CoolBox for .cool and .hic file visualization in the Jupyter nootebook.

About

Tools for DLO HiC data analyze

License:GNU General Public License v3.0


Languages

Language:Python 76.9%Language:Cython 10.6%Language:Mako 7.4%Language:JavaScript 3.0%Language:Dockerfile 1.2%Language:CSS 0.9%