search

sansomlab / cellhub

Python pipelines for large-scale multimodal single-cell data analysis on HPC clusters
MIT License
6 stars 2 forks source link

bug fixing r1-length and r2-length parameters in .yml and pipeline code #49

Closed nickilott closed 1 year ago

nickilott commented 1 year ago

A couple of bugs found when running cellranger count and speficying r1-length: "false" and r2-length: "false" in pipeline_cellranger.yml.

The issue is that iotools.py from cgat-core is used to process the PARAMS dict. The str2val function converts anything that is for example "false" to False. This resulted in the string "False" being substituted into the commandline status when the lengths were specified as "false" rather than "".

Also the pipeline looks for r1-length and r2-length in the gex: section of the yml rather than the count section so have copied this into the gex: section. Can probably remove from the count: section although haven't tested if it breaks anything downstream yet...

snsansom commented 1 year ago

Hi Nick, thanks - have tweaked slightly and fixed for vdj analysis also. For GEX and ADT I prefer to keep the option in the "count" section of the yaml as it applies to both. Let me know if this works for you and I will merge.

nickilott commented 1 year ago

Great, yes thanks. Sorry I left the print statement in :(