search

navinlabcode / copykat

Other
204 stars 54 forks source link

infer subpopulation in more than 100000 cells #53

Closed rained1989 closed 2 years ago

rained1989 commented 2 years ago

Hi, Thanks for the nice tool. I would like to cluster more than 100000 cells with CNA matrix (~120000 rows and 100000 column). But the matrix is too big for clustering. Should I filter some rows (abspos) with small variance?

Thanks.

gaobio commented 2 years ago

Yes, you can extract a subset of highly variable CNVs to do clustering.