Closed LTLA closed 8 years ago
For the batch effect, we could probably have a different sigma for each gene.
Ah, it's probably fine. That's just for the simulations, anyway, the framework doesn't require the distribution of delta (the batch effect variable) to be the same for every gene, only i.i.d. between batches.
For the batch effect, we could probably have a different sigma for each gene.