Way to pass lognormalized-regress out expression matrices

[Rafael-Silva-Oliveira]

Hey!

I'd like to try and do a DEA using an expression matrix that 1) has been lognormalized and 2) has had a variable regress out (different organs, so this needs to be regress out) which will result in some negative values. Any idea on how to make pydeseq2 work with negative values? Perhaps regressing out the variable I need to regress out with a contrained regression so that values don't become negative?

Appreciate any insights on this!