Questions About Fst, Dxy, and Genome-wide π Calculations

[xxxicyixxx]

Hello everyone,

I’ve encountered some issues while using the software. They might seem simple, but I’m unable to resolve them on my own, so I would greatly appreciate your help.

1. Based on average fst across windows #51 and Output Fst variance components #54, I would like to confirm whether the avg_hudson_fst in the results file pixy_fst.txt is calculated using the "average of ratio" method.
2. How are the Fst and dxy values between species pairs computed? Should they be calculated as a direct average of the avg_hudson_fst or avg_dxy values in the results file, or should they be computed using the formula below?
   (window 1 count_diffs + window 2 count_diffs) / (window 1 comparisons + window 2 comparisons)
   3.How should the genome-wide π value for a species be calculated?

Thank you in advance for your guidance!!!

[ksamuk]

1. Yes, its the sum of all the numerators and denominators of all of the individual FST ratios.
2. I'm not sure what you are asking here. Fst and Dxy are usually comparisons between two populations/species.
3. You can sum of all the counts of differences and comparisons for all the windows, as you mentioned in your second point. https://pixy.readthedocs.io/en/latest/output.html#post-hoc-aggregating. You can do this for any number of windows (e.g. all of them to get the total genome wide average).