[adegenet-forum] FW: SNP positional information

[Emmanuel Paradis]

Joanne,

Your problem is not so easy. One issue is that read.loci() in pegas does
not consider alleles as ordered, ie, A/B and B/A are the same (and B/A
is eventually reordered). I guess this is problematic with your data
because all "sites" are linked. So in your file, if a row is:

A/B A/B A/B

it means that all A's are on one chromosome and all B's are on the other.

I think it is worth for read.loci to have an option controlling whether
some loci are linked, and so treating alleles are ordered. Then a
function would extract the chromosomes of each invididual.

> From: adegenet-forum-bounces at lists.r-forge.r-project.org [adegenet-forum-bounces at lists.r-forge.r-project.org] On Behalf Of Joanne Berghout, Miss [joanne.berghout at mail.mcgill.ca]
> Sent: 14 April 2010 00:22
> To: adegenet-forum at lists.r-forge.r-project.org
> Subject: [adegenet-forum] SNP positional information
> 
> Hello adegenet users,
> 
> Is there a way that I can assign genetic positions to my SNPs?
> 
> I've just downloaded the ape, adegenet and pegas packages in R and
> I'm trying to analyze a data set of 400 individuals (13 populations)
> for 78 SNPs.
> 
> The data was collected by sequencing a single gene across each of the
> exons and looking for polymorphic sites both manually (chromatogram 
> inspection) and using PhredPhrap. I've been able up upload the data
> as a data frame (using read.loci) where each row is an individual and
> each column is a SNP (alleles coded as A/A). I would like to look at
> the haplotype differences in the different populations, as well as do
> some neutrality tests (specifically Tajima's D) and a few other
> standard descriptive statistics.
> 
> I am trying to avoid creating FASTA files of the whole sequence for 
> each of my individuals as I have multiple non-overlapping amplicons
> for each individual which (unless you also have a suggestion here)
> means that I would have to manually combine thousands of sequence
> reads (and edit for accuracy as PhredPhrap miscalled or just missed
> on quite a number of SNPs).

In case you did that, having the SNPs is simple with:

x[seg.sites(x)]

You may have also have a look at other packages dealing with SNP and/or
haplotypes: there are a few on CRAN.

Cheers,

Emmanuel

> 
> thanks, Joanne
> 
> PS. I'm a pretty new R user, though I have a fair bit of experience 
> with the R/qtl package as all my work so far has been on very 
> straightforward inbred mouse crosses. So, please, I would appreciate
> a little simplicity and explanation...
> _______________________________________________
> 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
> 

-- 
Emmanuel Paradis
IRD, Montpellier, France
   ph: +33 (0)4 67 16 64 47
  fax: +33 (0)4 67 16 64 40
http://ape.mpl.ird.fr/