[adegenet-forum] Different scatterplots each time (DAPC scatterplot)

[Jombart, Thibaut]

Hi Kaitlin,

PCA and DAPC solution will not change except for the sign of the axes, which is arbitrary. Only the distances between the points (individuals) matter.

The only possible stochastic thing here is if you use k-means clustering (find.clusters), which uses random starting states and therefore will not necessary give you the same results all the time (though qualitative changes are unlikely).

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 Kaitlin Marie Watt [kwatt22 at uwo.ca]
Sent: 27 January 2015 23:38
To: adegenet-forum at lists.r-forge.r-project.org
Subject: [adegenet-forum] Different scatterplots each time (DAPC scatterplot)

Hello,
 I’ve made a DAPC scatterplot using my microsatellite genotypes. However, when I run through the exact same code with the exact same parameters (No. clusters, No. of retained PCs, No. DFs), I get slightly different scatterplots each time (however, they are all similar in that none of them show much differentiation between clusters). Is this to be expected? And would someone please be able to explain (statistically) why I am getting slightly different results each time?

Thank you in advance!
Kaitlin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20150128/37e56e83/attachment.html>