Dear BioStars community:
I would like to know if there is a software or script that calculates Hudson's Fst estimator (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1205159/) at both Population and SNP level, the first to asses genetic distance between population and the second to detect highly divergent SNPs. I have used the EIGENSOFT software but this only produces Fst values at population level (pairwise matrix), but not for single SNPs. I also have used the well known method from Weir and Cockerham implemented in PLINK, but Bhatia et al, 2013 (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3759727/#B16) recommends to use Hudson's estimator because it behaves better with uneven population sizes.
Thanks in advance.
Using a VCF file as input, SNiPlay implements VCFtools that calculates Fst (Weir and Cockerham’s) for each marker... http://sniplay.southgreen.fr/cgi-bin/analysis_v3.cgi
Hi there,
Any updates on this? Did you find something useful? Or implement the estimator yourself? I found this: https://bio.cst.temple.edu/~hey/program_files/Sites/SITES_Documentation.htm but it doesn't look well maintained and I am pressed to find actual usage examples in the documentation.
I'm thinking about implementing a simple version myself.
Hope to hear your thoughts.
The second paper you mentioned actually links to a software package called EIGENSOFT. It is available here: http://www.hsph.harvard.edu/alkes-price/software/. I have to admit that this has not made things any easier for me as I do not understand what a block-jackknife estimator is and there is no mention of Hudson estimators anywhere in the documentation (or source code). I'll continue keeping you posted.
I was also frustrated by the lack of detail in the documentation--I see no mention of the type of Fst estimator.
Try this http://spsmart.cesga.es/
Log in to answer this question.
