Sergei Pond

Dear @ArtPoon, Thanks for this highly detailed issue report. The strange behavior arises because the Slatkin-Maddison analysis has been ported to use CLI keyword arguments: ``` hyphy % hyphy sm...

Dear @yurivict, Should be resolved with version 2.5.43. Best, Sergei

Dear @yurivict, Well, that was embarrassing. I forgot to merge the pull request fixing the issues before tagging off a 2.5.43. I've re-released 2.5.43 with the fixes included. Sorry about...

Dear @yurivict, Not normal. Unit tests should not take more than a second or two. I do not see this behavior with the 2.5.43 on the test system I have...

Dear @yurivict, You are 100% right. The hardcoded `/bin/bash` should have never been there. I fixed this issue in the 2.5.43 tag by replacing `/bin/bash` with a path that's guaranteed...

Dear @bonfaceonyango, It's whatever the branches are scaled in. Typically, expected substitutions/site/unit time. So 0.04 would be ~4%. Best, Sergei

Dear @kavithamallayaa, Can you be a bit more specific? Are you interested in site-level estimates, branch-level estimates, etc? For example, you can use RELAX in group mode to fit different...

Dear @kavithamallayaa, There are several things you could try. 1. You could simply estimate the dS tree (synonymous divergence); this will give you a branch-level average effect (`FitMG94` local option)...

Dear @kavithamallayaa, Can you tell me just a little bit more about what type of test you are most interested in? 1. Do you want to simply examine whether dS...

Dear @kavithamallayaa, We can definitely continue by e-mail. In the meantime, I suggest you try RELAX (with the recent release, `2.5.23`) which includes SRV. See https://github.com/veg/hyphy-analyses/tree/master/SRV If you suspect heterotachy,...