Running on asymetric MSA

[PabloGalazDavison]

Hi, this work is really cool. I was trying to run it on asymetric MSA, having one of them a select group of curated sequences (82) and the other, every other possible isoform for the same species (350). As I read in the preprint, this should be possible, yet the code evaluates and fails when the length of both MSA are different.
Can you please provide some insights into how to run a code with such characteristics please.
Thanks in advance.

[ulupo]

Hi @PabloGalazDavison! Unfortunately, I have not been notified about this issue, hence the slow reply. Indeed, we have not yet had a chance to push our internal code for handling these cases. It does something very simple: whenever, for a certain species, chain A has n sequences, and chain B has m sequences, if m > n then it adds m - n dummy sequences that are purely made of gaps to the chain A MSA, and if m < n then it adds n - m dummy sequences to the chain B MSA.

I'm happy to share code for this with you ahead of a proper push to the repo. Feel free to connect via email.