Hi!
The modkit dmr compare-tx-sites seems to swap samples. It reports the counts of sample a for sample b and vice versa. This reverses also effect size and log2 fold change.
May it be that these lines cause that?
let cond_a_counts = res.isoform_counts.pop().unwrap();
let cond_b_counts = res.isoform_counts.pop().unwrap();
GeneDmrScore
Hi!
The modkit dmr compare-tx-sites seems to swap samples. It reports the counts of sample a for sample b and vice versa. This reverses also effect size and log2 fold change.
May it be that these lines cause that?
GeneDmrScore