Saturation concerns

[Jessica Maccaro]

Hi there, 

I have a more conceptual question: 

I ran the branch site model using ete3 evol on single copy orthologs I have for 5 species. Several areas of positive selection came up but the AvgKs value appears quite large ( most of them greater than 1). Does this matter for this type of analysis, specifically using nucleotides rather than amino acid sequences? Should I scrap these, or should I consider them reliable if the alignments %ID is say greater than 70%? 

Thanks, 

Jess