[adegenet-forum] dealing with colinear variables in dapc

[Jombart, Thibaut]

Hello, 

collinearity is not the problem here, but variables with NULL variance give undefined F statistics and should this be discarded.

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 sophie bouchet [bouchet.sophie at free.fr]
Sent: 30 July 2013 17:00
To: adegenet-forum at lists.r-forge.r-project.org
Subject: [adegenet-forum] dealing with colinear variables in dapc

Dear Thibault,

When running dapc on a dataset containing 174 individuals and 10000 SNP markers, I obtain this error message :

Erreur dans lda.default(x, grouping, ...) :
  variables 1 2 appear to be constant within groups

Do I need to remove colinear markers from the dataset before running dapc?

Thanks,

Sophie