I understand Ohana estimates the best variance value in the range of a and a+h, when doing a branch specific selection scan. How should I choose h? Would providing an h of 10 or 100 make a difference? It seems a+10 would be a narrower space for interpolation than a+100. Moreover, the nemeco tree for the -cs matrix also looks different when I compare a+10 vs a+100. In the later the focal population looks more stretched.
Thank you.
I understand Ohana estimates the best variance value in the range of a and a+h, when doing a branch specific selection scan. How should I choose h? Would providing an h of 10 or 100 make a difference? It seems a+10 would be a narrower space for interpolation than a+100. Moreover, the nemeco tree for the -cs matrix also looks different when I compare a+10 vs a+100. In the later the focal population looks more stretched. Thank you.