[adegenet-forum] sPCA to infer dispersal

[Jombart, Thibaut]

Dear Rob, 

"local structures" (sensu sPCA) arise when neighbouring individuals are more genetically different than expected at random. Dispersal could indeed result in such patterns.

Cheers
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 Rob Denton [denton.66 at osu.edu]
Sent: 13 May 2014 18:07
To: adegenet-forum at lists.r-forge.r-project.org
Subject: [adegenet-forum] sPCA to infer dispersal

Fellow adegenet users,

I am currently exploring using adegenet to quantify dispersal in a polyploid salamander. Because of its ploidy (3n), more traditional software for genetic assignments (such as geneclass, Bayesass) are not helpful.

I understand that the the local eigenvectors constructed in a sPCA summarize local patterns in genetic data, but I'm trying to understand if any aspect of this procedure can act to infer dispersal. These animals are associated with discreet ponds, so there exists an a priori population assignment. Can I interpret those local eigenvalues as proxies for dispersal?

Thanks in advance for any guidance,

Rob Denton