iarcbioinfo / codex-nf

Pipeline for copy number variant calling from Whole Exome sequencing data using CODEX

CODEX-nf

Pipeline for copy number variant calling from Whole Exome sequencing data using CODEX

Prerequisites

• R with package CODEX
• Rscript

and the following files:

• a bed file with the positions to consider
• a folder with Normal BAM files
• a folder with Tumor BAM files

Usage

nextflow run iarcbioinfo/CODEX-nf --dirNormal Normal --dirTumor Tumor --bedFile positions.bed --outdir output

Output

A series of files projectname_chromosome_optK_mode.txt, where chromosome takes its values in [1,2,...,22,X,Y], while projectname, optK, and mode are the same for all files. projectname and mode are specified by the user with options --project and --mode, respectively, and optK is computed by package CODEX.

All parameters

PARAMETER	DEFAULT	DESCRIPTION
--help	null	Print usage and parameters
--dirNormal	Normal	Path to Normal BAM files
--dirTumor	Tumor	Path to Tumor BAM files
--outdir	.	Path to output directiry
--bedFile	positions.bed	Path to bed file with positions to consider
--rem_from_bed	_random|chrUn|GL000209R|GL000191R|GL000194R	Strings to exclude from the bed file chromosome list
--project		Project name
--mem	5	Memory requested
--cpus	1	CPUs requested
--seg_mode	fraction	Mode for the segmentation algorithm (fraction or integer)

codex-nf's Issues

Nextflow process CODEX_findoptK_allchr repeats the same operation 24 times

It would be better to do it just once and then output the result to the next process, but currently, doing so stops the parallelisation across chromosomes (the segmentation process is only luanched on 1 chromosome).

All chromosomes must be present in the bed file

It would be better to check which chromosomes are in the bed file, and then run CODEX only on these chromosomes