Skip to content

ding-lab/LR_SV_pipe

Repository files navigation

A pipeline for calling SVs from long-read sequencing data

This pipeline is designed for calling somatic structural and small variants from long-read sequencing data (e.g. Pacbio-HiFi and Ont). It works on LSF job scheduler and can run multiple jobs in parallel. Basically, this pipeline is composed of two steps: LR_0_StrucVar.pl for structral variants, and LR_1_SmallVar.pl for small variants.

Usage: a. perl LR_0_StrucVar.pl config4StrucVar.tsv b. perl LR_1_SmallVar.pl config4SmallVar.tsv

Prerequisite

Option 1: set a environment for LSF job on compute1 by adding the following to ~/.bashrc file:

export PATH=/rdcw/fs1/dinglab/Active/Projects/yuweiz/anaconda/envs/longread_cv/bin/:$PATH

export STORAGE1=/rdcw/fs1/dinglab/Active export SCRATCH1=/scratch1/fs1/dinglab/Active

export LSF_DOCKER_VOLUMES="$STORAGE1:$STORAGE1 $SCRATCH1:$SCRATCH1"

do NOT forget to run source ~/.bashrc

Option 2: modify the configure file (config.tsv) to specify the location of the softwares (minimap2 & Sniffles2) and the minimap2 index (if available) in the configure file. for example, MINIMAP2 ${your_path}/minimap2

Step 1

prepare your sample list file (sample.lst)

  format: #id    sequencing_platform    path_of_fastq

Note: specify the sequencing platform as the parameter -x used in minimap2 (https://lh3.github.io/minimap2/minimap2.html).

Step 2

modify the parameters configure file (config.tsv), inclduing the paths of output "OUTDIR", sample list "SAMPLE", softwares and index and the bsub setting.

Step 3

run perl LR_0_StrucVar.pl config4StrucVar.tsv

Please be sure the output dir is writtable and all softwares can be invoked.

Take care of the Sniffles mode!!! This pipeline defaultly runs both basic and mosaic (for low-frequency/non-germline SVs) modes.

Step 4

Small variant calling is based on ClairS (https://github.com/HKU-BAL/ClairS). This step requires the output bam file from Minimap2!

a. load new environment:

export PATH=/rdcw/fs1/dinglab/Active/Projects/yuweiz/anaconda3/envs/clair3/bin:$PATH

b. set up a configure file and a sample list.

  sample list format: #id    tumor_bam_path    normal_bam_path (if none, will run in the tumor only mode)

c. run perl LR_1_SmallVar.pl config4SmallVar.tsv

Contact

Yuwei ZHANG [email protected]

About

Call SVs from long-read sequencing data

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages