search

satijalab / sctransform

R package for modeling single cell UMI expression data using regularized negative binomial regression
GNU General Public License v3.0
203 stars 33 forks source link

Sctransform with multiple samples per batch #157

Closed ConDem94 closed 1 year ago

ConDem94 commented 1 year ago

Hi,

Thank you so much for this amazing normalization tool. I would like to ask, I am currently trying to integrate 5 studies (batches) with varying number of samples (pancreatic cancer). I saw that some samples have lower sequencing depth than others following my QC filtering step.

Therefore, should I perform sctransform per sample, per batch or the merged dataset (all 5 studies, 68 samples).

Thank you!

BW, Constantinos

saketkc commented 1 year ago

We recommend running sctransform per sample followed by integration.