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

Roman Luštrik roman.lustrik at biolitika.si
Mon Apr 3 09:57:02 CEST 2017


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


More information about the adegenet-forum mailing list