[adegenet-forum] (no subject)

[Thibaut Jombart]

Dear Axel,
> Hi,
>
> spca seems to work well on my AFLP data,
>
> 3 questions :
>
> -  I need to scale (scale =TRUE) the data, and I do not understand the reason.
>   
Why do you 'need' to scale? Is there an error otherwise? In any case, 
scaling may be useful, but it is never mandatory.
> - Another problem  the function screeplot does not work on my spca object.?
>   
What is the error message?
> - Is it possible to start spca from a  matrix of genetic distance between genotypes ?
>   
Not currently, and likely not in a near future. I would have to check if 
the principle of sPCA is extendable to distances matrices. What may be 
available sooner would be data transformation so that the Euclidean 
distances between transformed genotypes would correspond to a given 
genetic distance between raw genotypes.

Could you please post a sample dataset that reproduces the error? For 
instance, if 'obj' is your object, give this a try:
##
samp <- sample(1:nrow(obj),20)
objToPost <- obj[samp, 1:30]
##

if spatial coordinates are inside obj at other$xy, they should be subsetted 
at the same time (check objToPost$other$xy).

Also, we would need to know which version of adegenet you are using 
(there are changes between the devel and the last stable version).

Best regards,

Thibaut.


> If anyone can help ...
>
> Alex Baumel
>
> Institut Méditerranéen d'Ecologie et de Paléoécologie
> Bâtiment Villemin
> Europole de l'Arbois - BP 80
> F 13545 Aix-en-Provence cedex 04
> phone: (33) 442 908 453
> fax: (33) 442 908 448
>
> web page : http://alexbaumel.blogspot.com/
> _______________________________________________
> adegenet-forum mailing list
> adegenet-forum at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/adegenet-forum
>
>
>
>   


-- 
######################################
Thibaut JOMBART
CNRS UMR 5558 - Laboratoire de Biométrie et Biologie Evolutive
Universite Lyon 1
43 bd du 11 novembre 1918
69622 Villeurbanne Cedex
Tél. : 04.72.43.29.35
Fax : 04.72.43.13.88
jombart at biomserv.univ-lyon1.fr
http://biomserv.univ-lyon1.fr/%7Ejombart/
http://adegenet.r-forge.r-project.org/