Using Scientific Tools with Singularity
Some container tools are installed as .sif files on Alma and some are on dockerhub, where they can be used either locally with docker, or on Alma via singularity. These examples have been built at the ICR in order to simplify the usage of the complex environments that are required for them.
Note - singularity is only installed on the compute/interactive modes. It is not available on the login nodes.
bcl-convert from illumina
The software is from illumina The image is on Alma, located at: /opt/software/containers/singularity/build/bcl.sif
To use it, you need to bind a path on scratch to the container. It will use that path to write out log files. The general command is (asking for the help):
singularity exec --bind /path/to/your/logs:/var/log/bcl-convert /opt/software/containers/singularity/build/bcl.sif bcl-convert --help
If I want to check it works for, call help and bind it to your home path:
singularity exec --bind ~:/var/log/bcl-convert /opt/software/containers/singularity/build/bcl.sif bcl-convert --help
pvactools
pVACtools can be difficult to install and run on Alma so we have created a singularity image located in the shared singularity cache space.
The software is built as a docker image but can be run through singularity on Alma. To get the help you can run:
singularity exec /data/scratch/shared/SINGULARITY-DOWNLOAD/tools/.singularity/pvactools_latest.sif pvacseq run -h
An example to call the singularity image is, where you have created an out6 direcory and have your example data in pvacseq_example_data:
singularity exec \
--bind ./out6:/out6 \
--bind ./pvacseq_example_data:/pvaseq_example_data \
/data/scratch/shared/SINGULARITY-DOWNLOAD/tools/.singularity/pvactools_latest.sif \
pvacseq run pvacseq_example_data/annotated.expression.vcf.gz HCC1395_TUMOR_DNA HLA-A*29:02,HLA-B*45:01,DRB1*04:05 all out6 -e1 8,9,10 -e2 15
--bind parameter.
pisca from the ICR
The image is on dockerhub, located at: docker://icrsc/pisca-run. It was created for Heather Grant by the RSE team in Scientific Computing.
Pull it first into the directory you want to run it from, and then run it with a call that binds the directory in which you have the xml files to the container. The general command is: