BEB and NEB

[sunny kevin]

Hello all,

I did a branch site models on 5000 orthologs (eukaryotes)

Around 500 orthologs, I'm able to find positive selection sites in the outfile(.mlc).

Do I have to do BEB and NEB analysis further ?

How do I provide the functional annotation for the orthologs ?

(Is their standalone tools which can do ?)

Suggestions appreciated.

Thanks

Kevin

[Sandra AC]

Hi Kevin,

You should ignore the NEB a analysis and focus on the results output by the BEB analysis (the latter replaces the former, see Yang et al. 2005 for more details about the BEB analysis). You should report only those sites that have a probability of 95% or 99% of being positively selected (those sites with a "*" or with "**" in the BEB output).

With regards to how estimated proportions for sites belonging to specific site classes and the estimated ω, please see this thread here.

Hope this helps!
Sandra