v3.6Anchor id1 id1
Getting StartedAnchor id2 id2
Known IssuesAnchor id3 id3
Additional InformationAnchor id4 id4
Info |
---|
This page contains a quick start guide for earlier versions of Parabricks that are still available but no longer directly supported. Please refer to the latest version for direct support. |
v3.6
Getting Started
Connect to compute client.
...
Code Block | ||||
---|---|---|---|---|
| ||||
# Use scratch file system for temp space
export SCRATCH1=/scratch1/fs1/${COMPUTE_ALLOCATION}
# Use Active storage for input and output data
export STORAGE1=/storage1/fs1/${STORAGE_ALLOCATION}/Active
# Mapping for the Parabricks license(s) is required
export LSF_DOCKER_VOLUMES="/scratch1/fs1/ris/application/parabricks-license:/opt/parabricks $SCRATCH1:$SCRATCH1 $STORAGE1:$STORAGE1 $HOME:$HOME"
export PATH="/opt/miniconda/bin:$PATH"
# Use host level communications for the GPUs
export LSF_DOCKER_NETWORK=host
# Use debug flag when trying to figure out why your job failed to launch on the cluster
#export LSF_DOCKER_RUN_LOGLEVEL=DEBUG
# Use entry point because the parabricks container has other entrypoints but our cluster, by default, requires /bin/sh
export LSF_DOCKER_ENTRYPOINT=/bin/sh
# Create tmp dir
export TMP_DIR=${STORAGE1}"/parabricks-tmp"
[ ! -d $TMP_DIR ] && mkdir $TMP_DIR |
...
Info | ||||
---|---|---|---|---|
| ||||
|
Known Issues
VQSR does not support gzipped files.
CNVKit
--count-reads
does not work as expected. A separate CNVKit Docker image can be used an an alternative to this option.
Additional Information
-
- You may need to adjust your cores (
-n
) and memory (-M
andmem
) depending on your data set. 1 GPU server should have 64GB CPU RAM, at least 16 CPU threads.
2 GPU server should have 100GB CPU RAM, at least 24 CPU threads.
4 GPU server should have 196GB CPU RAM, at least 32 CPU threads.
- You may need to adjust your cores (
You can run this interactive (-Is) or in batch mode in the general or general-interactive queues.
You will probably want to keep the GPUs at 4 and RAM at 196GB unless your data set is smaller than the 5GB test data set.
There is diminishing returns on using more GPUs on small data sets.
Replace
command
with any of thepbrun
commands such asfq2bam
,bqsr
,applybqsr
, orhaplotypecaller
.Please refer to official Parabricks documentation for additional direction.