[adegenet-forum] Warning message when performing PCA analysis using adegenet package

Thibaut Jombart thibautjombart at gmail.com
Mon Apr 3 10:37:08 CEST 2017


Hi there,

this is the expected behaviour when loci are non-polymorphic. Why
would you like to keep them? Especially in the context of a PCA, where
it will creates variables with a variances of zero, and therefore Inf
values when scaling?

Cheers
Thibaut

--
Dr Thibaut Jombart
Lecturer, Department of Infectious Disease Epidemiology, Imperial College London
Head of RECON: repidemicsconsortium.org
sites.google.com/site/thibautjombart/
github.com/thibautjombart
Twitter: @TeebzR
+44(0)20 7594 3658


On 3 April 2017 at 09:21, Roman Luštrik <roman.lustrik at biolitika.si> wrote:
> Can you post a small, reproducible example (http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) which we can look at?
>
> Cheers,
> Roman
>
> ----
> In god we trust, all others bring data.
>
> ----- Original Message -----
> From: "Thomas HAINAUX" <thomas.hainaux at unamur.be>
> To: "Roman Luštrik" <roman.lustrik at biolitika.si>
> Cc: adegenet-forum at r-forge.wu-wien.ac.at
> Sent: Monday, April 3, 2017 10:16:23 AM
> Subject: Re: [adegenet-forum] Warning message when  performing PCA analysis using adegenet package
>
> Not really, but i removed the spaces and it didn't change anything.
>
> Le Lundi 3 Avril 2017 09:57 CEST, Roman Luštrik <roman.lustrik at biolitika.si> a écrit:
>
>> Are the spaces around NA (NA.char = " NA ") intended?
>>
>> Cheers,
>> Roman
>>
>> ----
>> In god we trust, all others bring data.
>>
>> ----- Original Message -----
>> From: "Thomas HAINAUX" <thomas.hainaux at unamur.be>
>> To: adegenet-forum at r-forge.wu-wien.ac.at
>> Sent: Monday, April 3, 2017 9:47:38 AM
>> Subject: [adegenet-forum] Warning message when performing PCA analysis using  adegenet package
>>
>> Hi,
>>
>> I was performing an PCA analysis on a data set containing  4709 SNPs in
>> 144 individuals but I encountered some warning messages that i didn't understood.
>> I followed the instructions from the tutorial "An introduction to adegenet 2.0.0" (July 29, 2015).
>>
>>
>> 1) I opened my data and converted it to an genind  object and i got this warning message:
>>
>> > datacsv = read.csv2("170329_solcap_select_20_98_grouped_by_k3.csv ", header=TRUE)
>> > datagenind = df2genind(datacsv[,2:4709], sep= " | ", NA.char=  " NA ")
>> Warning message:
>> In df2genind(datacsv[, 2:4709], sep = " | ", NA.char = " NA ") :
>>   entirely non-type marker(s) deleted
>>
>> 2) Then I wanted to run the scalegen function and i got a second warning message.
>> > datasc = scaleGen(datagenind[,2:4709], NA.method= "mean")
>> Warning message:
>> In .local(x, ...) : Some scaling values are null.
>>  Corresponding alleles are removed.
>>
>> After running the scalegen function, the number of loci went from 4709 to 3694.
>> Why these loci are being removed? Is it possible to keep these value?
>>
>>
>> Thank you very much beforehand for your help and your time.
>> I apologise if my questions might be trivial for you, I just started using R recently and i'm not an expert yet.
>>
>>
>> Best regards
>> Thomas Hainaux - Researcher
>> URBV - Plant Biotechnology
>> Namur University
>> Rue de Bruxelles 61 - 5000 Namur
>> Belgique
>>
>>
>> _______________________________________________
>> 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
> _______________________________________________
> 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


More information about the adegenet-forum mailing list