GWAS studies with RADSeq

Question

[Picasa]

Dear Biostar's community,

I'm performing GWAS using RADseq. I have 30 samples, my goal is to discriminate these samples using SNP variation.

I've succesfully applied Stacks (http://catchenlab.life.illinois.edu/stacks/) with my data. And I got a vcf file at the end.

1) I've tried a PCA method using Plink and flashPCA but I didn't have so much result. I know that we have to LD prune our dataset before a PCA but I'm not sure about the others filtering method.

http://pngu.mgh.harvard.edu/~purcell/plink/thresh.shtml

Any recommendations about default Missingness per individual, Missingness per marker, MAF, Hardy-Weinberg equilibrium treshold ?

2) Can you recommend other methods than PCA ?

Answer 1

[alexisdereeper]

You can use SNiPlay that allows to manage VCF file and proposes methods for filtering, PCA, MDS plot, structure... http://sniplay.southgreen.fr/cgi-bin/analysis_v3.cgi