SLURM installation instructions

[dstueckm]

Hello and thanks for creating this pipeline!

I noticed that the two descriptions for setting up the snakemake pipeline are for local runs or lsf clusters. Would you be able to provide a description of how to set this up and run the pipeline on a cluster with a SLURM scheduler?

Thank you,
Daniel

[annavpo]

Hello @dstueckm ,
I have updated the instructions to use the pipeline on slurm: You would need to create a slurm profile first as prompted on READ.me step 7. Then rename CATD_snakemake/bin/config_slurm.yaml to CATD_snakemake/bin/config.yaml making it your main config file. Finally you need to add the time=get_time('minorOps') under resources in all Snakefiles for each Module. Example here:

rule proportionsInAdmixture_run:
input:
getC2(getC1(getBulks('{sample}'), '{sample}'),'{sample}')

output:
"Output/{sample}_res_proportionsInAdmixture.rds"

conda:
"env.yaml"

benchmark:
"Benchmarks/{sample}_proportionsInAdmixture_benchmark.txt"

params:
    markers=config['markers']

resources:
mem_mb=getMB(config['mem_gb']['minorOps']),
    time=get_time('minorOps')

shell:
"Rscript Modules/ADAPTS/proportionsInAdmixture_run.R {input} {params.markers} {output}"

This is a bit hacky/tedious but would work for now and I am working on updating the scripts to opt for slurm/lsf pipeline more easily. If you have any issues let me know!