GATK4 RNA-Seq Germline using Singularity Container

1 Jump start
2 Details
3 Log on to O2
4 Start interactive job, and create working folder
5 Load the pipeline related modules
6 Build some testing data in the current folder
7 Take a look at the example files
8 The original bash script
9 Sample output from the test run
10 Run the pipeline
11 Monitoring the jobs
12 Check job logs
13 Re-run the pipeline in case some jobs fail
14 Call variance:
15 To run the workflow on your own data

This page shows you how to run GATK4 using our recently installed Singularity GATK4 container. The runAsPipeline script, accessible through the rcbio/1.0 module, converts the bash script into a pipeline that easily submits jobs to the Slurm scheduler for you.

Features of this pipeline:

Given a sample sheet, generate folder structure for data processing
Submit each step as a cluster job using sbatch.
Automatically arrange dependencies among jobs.
Email notifications are sent when each job fails or succeeds.
If a job fails, all its downstream jobs automatically are killed.
When re-running the pipeline on the same data folder, if there are any unfinished jobs, the user is asked to kill them or not.
When re-running the pipeline on the same data folder, the user is asked to confirm to re-run or not if a step was done successfully earlier.

The workflows are downloaded from: https://github.com/gatk-workflows/gatk4-rnaseq-germline-snps-indels and modified to work on O2 slurm cluster.

Notice the original workflow uses reference and annotation files listed in this file:

https://github.com/gatk-workflows/gatk4-rnaseq-germline-snps-indels/blob/master/gatk4-rna-germline-variant-calling.inputs.json

We download the genome reference and all annotation files from:

https://console.cloud.google.com/storage/browser/genomics-public-data/references/Homo_sapiens_assembly19_1000genomes_decoy/ except for the gtf file, which is downloaded from here: https://console.cloud.google.com/storage/browser/gatk-test-data/intervals?project=broad-dsde-outreach

We then modified the json file to this one:

/n/shared_db/singularity/hmsrc-gatk/scripts/gatk4-rna-germline-variant-calling.inputs.template.json

Please modified the file to use your own reference and annotation files as you like.

Jump start

Here are the commands to test out the workflow using example data. The whole run need a few hours if the cluster is not busy.

ssh user123@o2.hms.harvard.edu

# set up screen software: https://wiki.rc.hms.harvard.edu/pages/viewpage.action?pageId=20676715
cp /n/shared_db/misc/rcbio/data/screenrc.template.txt ~/.screenrc

screen

srun --pty -p interactive -t 0-12:0:0 --mem 16000MB -n 2 /bin/bash

mkdir /n/scratch/users/${USER:0:1}/$USER/testGATK4
cd /n/scratch/users/${USER:0:1}/$USER/testGATK4

module load gcc/6.2.0 python/2.7.12 java/jdk-1.8u112 star/2.5.4a

export PATH=/n/shared_db/singularity/hmsrc-gatk/bin:/opt/singularity/bin:$PATH

# setup database. Only need run this once. It will setup database in home, so make sure you have at least 5G free space at home.
setupDB.sh

cp /n/shared_db/singularity/hmsrc-gatk/scripts/* .

buildSampleFoldersFromSampleSheet.py sampleSheet.xlsx

runAsPipeline fastqToBamGermline.sh "sbatch -p short --mem 4G -t 1:0:0 -n 1" noTmp run 2>&1 | tee output.log

# check email or use this command to see the workflow progress
squeue -u $USER -o "%.18i %.9P %.28j %.8u %.8T %.10M %.9l %.6D %R %S"

# After all jobs finish run, run this command to start database, and keep it running in background
runDB.sh &     

for json in unmappedBams/group*/*.json; do
	echo working on $json
    [ -f $json.done ] && continue
	java -XX:+UseSerialGC -Dconfig.file=your.conf -jar /n/shared_db/singularity/hmsrc-gatk/cromwell-43.jar run gatk4-rna-germline-variant-calling.wdl -i $json 2>&1 | tee -a $json.log && touch $json.done
done

# rerun the above command if anything does not work

# find all .vcf files and create sym links in final folder. If there are dulplicate files: 
mkdir finalVCFs
ln -s  $PWD/cromwell-executions/RNAseq/*/call-VariantFiltration/execution/*.variant_filtered.vcf.gz finalVCFs/ 2>/dev/null

#merge all CVFs to single VCF file
vcfFiles=`ls finalVCFs/*.variant_filtered.vcf.gz`
module load bcftools/1.9
bcftools merge -o final.vcf --force-samples $vcfFiles

# Stop database
killall runDB.sh

Details

Log on to O2

If you need help connecting to O2, please review the How to login to O2 wiki page.

From Windows, use MobaXterm or PuTTY to connect to o2.hms.harvard.edu and make sure the port is set to the default value of 22.

From a Mac Terminal, use the ssh command, inserting your HMS ID instead of user123:

Start interactive job, and create working folder

Load the pipeline related modules

Build some testing data in the current folder

Take a look at the example files

The original bash script

How does this bash script work?

There is a loop that goes through the two group folders and samples, convert the fastq files into unmapped bam files

These comments are recgnized by our pipeline runner and the command following them are submitted as slurm jobs:

This command will generate new bash script named slurmPipeLine.201907200946.sh in flag folder (201907200946 is the timestamp that runAsPipeline was invoked at). Then test run it, meaning does not really submit jobs, but only create a fake job id, 123 for each step. If you were to append run at the end of the command, the pipeline would actually be submitted to the Slurm scheduler.

Ideally, with 'useTmp', the software should run faster using local /tmp disk space for database/reference than the network storage. For this workflow, we don't need it.

Sample output from the test run

Note that only step 2 used -t 50:0, and all other steps used the default -t 10:0. The default walltime limit was set in the runAsPipeline command, and the walltime parameter for step 2 was set in the bash_script_v2.sh

Run the pipeline

Thus far in the example, we have not actually submitted any jobs to the scheduler. To submit the pipeline, you will need to append the run parameter to the command. If run is not specified, test mode will be used, which does not submit jobs and gives theplaceholder of 123 for jobids in the command's output.

Monitoring the jobs

You can use the command:

To see the job status (running, pending, etc.). You also get two emails for each step, one at the start of the step, one at the end of the step.

Check job logs

You can use the command:

This command list all the logs created by the pipeline runner. *.sh files are the slurm scripts for eash step, *.out files are output files for each step, *.success files means job successfully finished for each step and *.failed means job failed for each steps.

You also get two emails for each step, one at the start of the step, one at the end of the step.

Re-run the pipeline in case some jobs fail

You can rerun this command in the same folder

This command will check if the earlier run is finished or not. If not, ask user to kill the running jobs or not, then ask user to rerun the successfully finished steps or not. Click 'y', it will rerun, directly press 'enter' key, it will not rerun.

Call variance:

To run the workflow on your own data

To instead run workflow on your own data, transfer the sample sheet to your local machine following this wiki page and modify the sample sheet. Then you can transfer it back to O2 under your account, then go to the build folder structure step.

Let us know if you have any questions. Please include your working folder and commands used in your email. Any comment and suggestion are welcome!

HMS IT RC O2

NOTICE: FULL O2 Cluster Outage, January 3 - January 10th