[adegenet-forum] interpreting scatter plot from dapc analysis

[Jombart, Thibaut]

Hi there,

1) It is the plot of two principal components of the analysis (by default x=1 and y=2), i.e. the synthetic variables best discriminating your groups (in the sense of the ratio of variance between group / within group)

2) It is exactly what it reads on the box, i.e. a one dimensional representation of the chosen principal component (typically the first and only one); tick marks are individuals, and the density curves are derived by kernel density estimation, i.e.: they show you where individuals are.

The phrase:
"proportion of individuals in groups against the eigenvalues of the first discriminant function"
does not really make sense I'm afraid ;) It may be useful to check the paper, if you have not yet, hopefully this would bring some clarifications (apologies if not).

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 kin onn [kin_onn at yahoo.com]
Sent: 01 October 2015 19:11
To: adegenet-forum at lists.r-forge.r-project.org
Subject: [adegenet-forum] interpreting scatter plot from dapc analysis

Dear all,

I have a question regarding the interpretation of two types of scatter plots from the dapc analysis
1) the cluster plot from the function scatter(dapc1). The axes are not labeled and I was wondering what they were
2) the scatter that represents a single discriminant function from the function scatter(dapc1,1,1). In the tutorial it says that it plots the density of individuals against the discriminant function. Is this the proportion of individuals in groups against the eigenvalues of the first discriminant function? How should we interpret this graph?

Thank you in advance!


Best regards,
Chan
Biodiversity Institute and Department of Ecology and Evolutionary Biology
University of Kansas
Dyche Hall, 1345 Jayhawk Blvd
Lawrence, KS 66045-7593
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20151001/5f3cdd9a/attachment.html>