Skip to content

Genotyping

MSG

Note: this tool was installed on Ada and is not currently available on any cluster. Request it to be installed on a current cluster if you want to use it.

GCATemplates available: no

MSG homepage

Multiplexed shotgun genotyping (MSG).

A pipeline of scripts to assign ancestry to genomic segments using next-gen sequence data. This method can identify recombination breakpoints in a large number of individuals simultaneously at a resolution sufficient for most mapping purposes, such as quantitative trait locus (QTL) mapping and mapping of induced mutations.

First you will need to copy the MSG repository which has already been downloaded on Ada. The msg latest release v0.4.9 has bugs so use our copy.

In your working directory where your job script will be located, rsync the already downloaded msg repo:

rsync -r /software/hprc/Bio/MSG/f361369/ msg

You can copy the following sample files to your working directory to get an idea of what input files are needed and their formats and also you can run the sample job script without any making additional changes.

cp /software/hprc/Bio/MSG/example_msg_files.tgz ./
tar xzf example_msg_files.tgz

This is the sample Job Script run_msg_test_example.sh from the sample data (this is for the sample data set, adjust walltime as needed for your larger data files and update msg.cfg and update.cfg as needed for your project)

You can run the job script on the sample data without having to make any additional adjustments

sbatch run_msg_test_example.sh

This is the example job script run_msg_test_example.sh taken from running on Ada. If you plan to use MSG on Grace, please contact the HPRC helpdesk to find the proper modules to load.

#!/bin/bash
#SBATCH --export=NONE               # do not export current env to the job
#SBATCH --job-name=my_job           # job name
#SBATCH --time=7-00:00:00           # max job run time dd-hh:mm:ss
#SBATCH --ntasks-per-node=1         # tasks (commands) per compute node
#SBATCH --cpus-per-task=48          # CPUs (threads) per command
#SBATCH --mem=360G                  # total memory per node
#SBATCH --output=stdout.%x.%j          # save stdout to file
#SBATCH --error=stderr.%x.%j           # save stderr to file

module load Pyrex/0.9.9-intel-2015B-Python-2.7.10
module load Stampy/1.0.28-intel-2015B-Python-2.7.10
module load BWA/0.5.7-intel-2015B
module load pysam/0.2-intel-2015B-Python-2.7.10-SAMtools-0.1.9
module load Biopython/1.65-intel-2015B-Python-2.7.10
module load XZ/5.2.1-intel-2015B
module load libpng/1.6.21-intel-2015B
module load libjpeg-turbo/1.4.1-intel-2015B
module load PCRE/8.38-intel-2015B
module load Java/1.8.0_92
module load cURL/7.47.0-intel-2015B
module load R_tamu/3.3.1-intel-2015B-default-mt

perl msg/msgUpdateParentals.pl
perl msg/msgCluster.pl
Back to top