search

zavolanlab / SCINPAS

Single Cell Identification of Novel PolyA Sites.
Apache License 2.0
2 stars 0 forks source link

Is SCINPAS compatible with STARsolo generated BAMs? #7

Open SamBryce-Smith opened 9 months ago

SamBryce-Smith commented 9 months ago

Hi Youngbin and Dominik,

Really neat approach! We're interested in trying out SCINPAS on some public scRNA-seq datasets to detect expression of a pre-defined list of target PAS. We've processed our data using STARsolo, but all mentions in the manuscript and README refer to CellRanger processed BAMs. Have you ever successfully ran STARsolo processed BAMs through the SCINPAS workflow? If so, do you have any recommendations/advice on parameters/run modes prior to input with SCINPAS?

I'm aware that STARsolo has been designed to mimic the CellRanger outputs/workflow, but just wanted to check in before we give it a go.

Cheers, Sam

dominikburri commented 9 months ago

Hi Sam, Glad you are interested and want to try it. So far we didn't test it with reads processed by STARsolo, but I would think that it works. SCINPAS requires the CB and UR tags, so the corrected cell and UMI barcodes. It works on soft-clipped reads, so those need to be available. Given that CellRanger uses STAR for the mapping, I'm rather confident that reads mapped with STARsolo work too. Give it a shot and let us know if you run into issues.

Best, Dominik