Hi

I would like to compute dn/ds ratio for each gene in my genome and I am looking for a conceptual protocol on how to do it and which tools to use. For example:

First i need to find a homologue (ortholog ? ) to my gene in species A. In this case would it be better to simply take the closest species and then blast/blat its sequences and find that homolog or would it be better to use a set of species and find family groups and from it do the consequent computations? Once I did that what would be the next step?

Is there any best practice protocol for this computation ?

Any advice would e more than appreciated... Please help!!