Closed wzheng0520 closed 2 weeks ago
I would like to second this. It doesn't make sense to re-run all indexing processes for every single run when they are already part of the genome config file. This should also be cut down noticeably the execution time, and the carbon footprint!
Tagging @apeltzer. This would be good to add, I believe we've got this internally already.
I have opened a PR for this (see link above), I think we can close this issue.
Description of feature
Hi guys,
Right now, I found for cellranger index, simpleaf index and simpleaf tx2pgene, the only way we can provide the pre-built indices is through using
--cellranger_index
,--salmon_index
and--txp2gene
parameters. I am thinking about if we already have the default path of those pre-built indices, once we use our own igenomes.config file, those indices should be directly pull from there likefasta
,gtf
, orstar_index
right now in the pipeline.I think if it is possible to add following lines before we defined their channel in the
scrnaseq.nf
to allow the pipeline can capture the those information directly from
igenomes.config
if the--genome
we provided have those information.Sincerely Winnie Zheng