search

AlexsLemonade / alsf-scpca

Management and analysis tools for ALSF Single-cell Pediatric Cancer Atlas data.
BSD 3-Clause "New" or "Revised" License
0 stars 1 forks source link

Add --include-introns for cellranger workflow #76

Closed allyhawkins closed 3 years ago

allyhawkins commented 3 years ago

Closes #72. I added a parameter to the run-cellranger.nf workflow to allow for flexibility to use cellranger with snRNA-seq samples. To use the --include-introns flag you would implement it at the time of running the workflow by executing:

nextflow run cellranger-quant/run-cellranger.nf --include_introns:"--include-introns"

I don't think this is a permanent fix for this. Before we run all of our samples I think a better solution might be to grab the seq-unit from the scpca-library-metadata.tsv file and based on if the seq-unit == nucleus, then we would use the flag for --include-introns.

However, since we are still in the benchmarking phase, I thought it might not be the best approach to give up complete control. It could be useful to compare some of the snRNA-seq runs using cellranger with and without this flag rather than include it as a default immediately.

jashapiro commented 3 years ago

Should we automatically switch modes based on the sample? That is, if the seq_unit is nucleus, we use --include-introns but if it is cell we do not?

I just realized that you addressed this in your PR submission, which I obviously didn't read again this morning. I blame Monday. Apologies.

I still think it makes sense to set the option up as an input to the process to facilitate later changes, but I'm really sorry for just repeating something you already addressed.

allyhawkins commented 3 years ago

No worries at all. I do agree we will need to do a final update before we run everything to automatically use the --include-introns for snRNA-seq samples, assuming we that's what we end up deciding on. I moved it to be an input to the process to get us started.