The situation I am dealing with is this: I had a bunch of many-to-many orthologs identified by OrthoFinder
. Problem is that the orthogroups are defined at the level of all eukaryotes (the input includes species from all over that branch), so the orthogroups are not informative of the functional relationships between the sequences within. (This is what I am actually interested in).
Is there a way to figure out which members of the orthogroups share functional aspects, and which don't?
One idea I had was to get the sequences' domains annotated, and then compare the sequences in terms of domain composition (which domains and how many).
Another idea I had was to look at the PANTHER annotations for the sequences and see if they share the correct PANTHER sub-families (which I believe suggest shared functionality?).
But I am still confused, and I'm looking for some advice on this issue.
Have you considered running CDD searches on representatives of groups you have found?
Why
CDD
in specific?