Hello all,

I am trying to infer local adaptation from Qst-Fst comparisons. I have 4 populations, say popA, popB, popC, popD. Each of the populations have 6 different genotypes and each genotype has 3 biological replicates. In total I have 72 samples. After sequence assembly, I obtained around 30000 genes for my dataset. What I would like to do now is calculate the Qst values. I already have the pairwise Fst values (i.e.,

1. PopA-PopB
2. PopA-PopC
3. PopA-PopD
4. PopB-PopC
5. PopB-PopD
6. PopC-PopD

I have a couple of questions:

1) How is the Qst calculation usually done? Is it done pairwise (since I have pairwise Fst values) or should I do it for all populations at once?
2) Based on the formula implemented in most of the papers, I got the following impression (please correct me if I am wrong):

For the pairwise PopA-PopB, Qst can be calculated as:

Gb = Variance calculated with PopA and PopB (pooled)
Gw = (Variance of PopA + Variance of PopB)/2
Qst = Gb / (Gb + 2 * Gw)

Please help!