Hello, "I'm calculating Fst values for each SNP to investigate genetic differentiation between populations. I'm using the Popgenome package for fixation index calculation. Typically, Fst is calculated only for biallelic sites. However, if a site is triallelic with frequencies of A=56%, T=43.5%, and C=0.5%, excluding samples with the C allele from the calculation could pseudo-biallelically represent the site. Would this approach be reasonable? Many analyses filter SNPs based on minor allele frequency (MAF), but would this be effective for Fst calculation as well?"

Thanks,