[adegenet-forum] Stepwise forward discriminant analysis

[Jombart, Thibaut]

Hello, 

I would recommend cross-validation, now available in the latest release of adegenet. See function xvalDapc.

Best

Thibaut

________________________________________
From: adegenet-forum-bounces at lists.r-forge.r-project.org [adegenet-forum-bounces at lists.r-forge.r-project.org] on behalf of Muriel Thomasset [thomassm at tcd.ie]
Sent: 24 May 2013 14:31
To: adegenet-forum at lists.r-forge.r-project.org
Subject: [adegenet-forum] Stepwise forward discriminant analysis

Dear all,

I’m using over 300 SNPs to discriminate 5 different populations. I’m doing DAPC and looking at alleles contribution which is very useful. I would like to reduce the number of markers to use and so find the best combination that would show the best discrimination.  I was thinking to do a stepwise forward discriminant analysis.

 I was wondering if someone has experience to do so on molecular markers?  For examples which package or functions to use?



Best regards



Muriel


--
Thomasset Muriel
Trinity College
Dublin 2, Ireland
+353858269170