[adegenet-forum] DAPC "reduced" confidence ellipse

[Virgilio Massimiliano]

many thanks I’ll give it a try :-)
all the best
M.

__________________________________
 Massimiliano Virgilio, PhD

Royal Museum for Central Africa

Joint Experimental Molecular Unit (JEMU)

Leuvensesteenweg 13, B-3080 Tervuren, Belgium, +32 (0) 27695366

massimiliano.virgilio at africamuseum.be<mailto:massimiliano.virgilio at africamuseum.be>

http://www.africamuseum.be/home/contact/staff/VIRGILIO_Massimiliano/

http://jemu.myspecies.info







On 25 Feb 2015, at 13:56, Jombart, Thibaut <t.jombart at imperial.ac.uk<mailto:t.jombart at imperial.ac.uk>> wrote:

Hi Massi,

tough one! For this kind of customisation you'd have to calculate the ellipse manually. To figure out how it's done in ade4, check out the code of s.class, and scatterutil.ellipse.

Cheers
Thibaut
________________________________
From: adegenet-forum-bounces at lists.r-forge.r-project.org<mailto:adegenet-forum-bounces at lists.r-forge.r-project.org> [adegenet-forum-bounces at lists.r-forge.r-project.org<mailto:adegenet-forum-bounces at lists.r-forge.r-project.org>] on behalf of Virgilio Massimiliano [massimiliano.virgilio at africamuseum.be<mailto:massimiliano.virgilio at africamuseum.be>]
Sent: 25 February 2015 10:54
To: adegenet-forum at lists.r-forge.r-project.org<mailto:adegenet-forum at lists.r-forge.r-project.org>
Subject: [adegenet-forum] DAPC "reduced" confidence ellipse

good morning,

I have a nice scatter graph of my dapc:

scatter(dapc348_country,
        col=myColCountry,
clabel=.5,
        bg="grey",
        cellipse=2.5,
        scree.pca=TRUE, ratio.pca=0.15,
        scree.da=TRUE, ratio.da=0.15,
        cstar=1,  axesell=FALSE,
        cex=1, grid=FALSE, addaxes=TRUE)

what I would like to do is to re-draw one of the confidence ellipses by excluding a few outliers from one of the populations and possibly show all ellipses calculated for each population + the “reduced” ellipse calculated on the population with excluded outliers

I’m not sure this is really possible though…

many thanks in advance for your answer

all the best

Massi

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20150226/38fa964b/attachment.html>