Dear Sergei,

As you suggested based on your instruction, I tried to run "Standard Analyses >Psoitive Selection>CompareSelectivcePressureVL.bf" on two multiple alignments (two populations). but, I didn't find the sites evolving differentially at P <= 0.05. In order to confirm my result, I would like you to judge if my work is valid (there is no clear instruction for this):

1). Prepared two nex files with Pyphy, including nucleotide multiple alignments and inferred trees. Alignments were aligned by MEGA and tested on Datamonkey server (no stop codons) and trees were inferred by Neighbour-Join method with PyPhy

2). Modes for alignments are both 012232 (tested by Pyphy--NucModeCompare.bf).

Please let me know if something is wrong with my works!  If not, I have some questions for the result?

I also did SLAC, FEL, and IFEL on two populations and found that many sites under positive selection are different for two populations. And even the patterns of positive selection (P<0.05; SLAC, FEL and IFEL) are also different. How to explain it (plus no sites found evolving differentially at P <= 0.05 in two populations)?

Thanks!

Ben

Dear Sergei,

Thank you for the resposne!  I did try to re-run my data using another comparison method: "CompareSelectivePressure.bf" and got the same results as one I run before. 

I have two questions: (if the approaches and methods I used are correct)

1). Does it mean that at population level, there is no differential selection between two populations?
2). How to explain the differences of selection patterns (eg. the sites  are positively selected differently by SLAC, FEL, IFEL when comparing two populations)? --- if it is within-hosts, transient subtitutions (in tracellular, intra-individual, etcs) are different in two populations?  or others!  Please give me more precise explainations!

Thanks!

Ben