search

wheaton5 / souporcell

Clustering scRNAseq by genotypes
MIT License
168 stars 46 forks source link

How to re-run clustering with singularity install? #10

Closed chris-mcginnis-ucsf closed 5 years ago

chris-mcginnis-ucsf commented 5 years ago

Hello again!

Potentially naive question here -- If I ran SoupOrCell using the suggested singularity install, how can I run specific parts of the pipeline as you describe in the 'hard-install' section? I just applied the pipeline to a new dataset but realized I set the wrong k and want to just re-do the clustering step.

Thanks again for the new tool -- it is extremely useful!!!

Chris

wheaton5 commented 5 years ago

You should be able to delete the .done file for the step you want to rerun and rerun the souporcell_pipeline.py script (with new parameters). But I would recommend that if you rerun one part, you should rerun everything downstream. So if you rerun clustering, you should rerun doublet detection and ambient rna estimation/genotyping. So that will be clustering.done, troublet.done, consensus.done.

Also you can run these tools individually similar to my hard-install section except through the singularity container (I should update the docs more). All you have to do is run singularity exec souporcell.sif command where command might be souporcell.py or troublet (you can see the examples I have in the hard install section, but using the singularity container it will be no installation!)

Also its all lowercase souporcell lol. And pronounced as "supercell" just spelled differently lol.

chris-mcginnis-ucsf commented 5 years ago

Sweet! Will let you know if I run into any issues.

And duly noted -- we here in the Bay Area are big fans of extraneous capitalizations :)

wheaton5 commented 5 years ago

I will close this for now. Feel free to reopen if you had any issues.