Pre-processing Seurat object is faster when FindVariableFeatures runs before NormalizeData and ScaleData

Question

Pre-processing Seurat object is faster when FindVariableFeatures runs before NormalizeData and ScaleData

MahdiMoradiMarjaneh opened this issue 5 years ago · 1 comments

MahdiMoradiMarjaneh commented 5 years ago

Currently ScaleData in doubletfinder function takes too long for medium to large datasets. The run time would be significantly shorter if FindVariableFeatures runs before NormalizeData and ScaleData.
https://satijalab.org/seurat/v3.0/mca.html``

Answer 1 · 2020-05-27T14:36:00.000Z

also running run_doubletfinder() rather than find_singlets() speeds the scaling process up.