A community effort to collect a curated set of analysis pipelines built using Nextflow.

nf-core

Hi,
<@U0628B1GNFP>, most recent member of <#C05V9FRJYMV|> will be working into adding BBMAP/BBCPLIT in <#CGFUX04HZ|>.
And we were just wondering what is your use case in <#CE8SSJV3N|> for it, and how do you differentiate the 2 genomes

It's used to filter out anything that maps preferentially to contaminant genomes. Some docs here: <https://nf-co.re/rnaseq/3.18.0/docs/output/#bbsplit>.

The <https://github.com/nf-core/modules/blob/master/modules/nf-core/bbmap/bbsplit/main.nf|module> is called from <https://github.com/nf-core/modules/blob/master/subworkflows/nf-core/fastq_qc_trim_filter_setstrandedness/main.nf|the preprocessing subworkflow>, which only retains the 'primary'-mapping reads for downstream analysis.

I'm not fond of BBMap, it's not what I'd call high quality software. I had to do some <https://github.com/nf-core/modules/blob/788825f9e429b03b892a1fbe7e8d25cce909a998/modules/nf-core/bbmap/bbsplit/main.nf#L74|pretty appalling hackery> to stop it continuously re-indexing, and it's still a beast for resource usage.

Thanks Jonathan. I agree that it consumes a lot of resources. When I was using it in the rnaseq pipeline I had to increase a lot the default resources up to 24 cpu, 90GB and 60h. But apart from that everything worked properly, probably thanks to the hackery you are mentioning!