Related to #161, is it still necessary to manually ensure that the Pearson residuals are non-negative in order to use SCTransform-normalised data for LIGER?
Do I just start from RunOptimizeALS after running SCTransform? And would I run SCTransform with all batches combined or for each batch separately?
I would also like to know whether I can use SCTransform on my data before starting with the Seurat-to-LIGER pipeline, and whether I should do it on samples separately or combined!
Related to #161, is it still necessary to manually ensure that the Pearson residuals are non-negative in order to use SCTransform-normalised data for LIGER?
Do I just start from
RunOptimizeALS
after runningSCTransform
? And would I runSCTransform
with all batches combined or for each batch separately?Is this tutorial up-to-date for the current version of LIGER?
http://htmlpreview.github.io/?https://github.com/satijalab/seurat-wrappers/blob/master/docs/liger.html
Thanks!