[adegenet-forum] Analysing mixed ploidy datasets

[Zhian Kamvar]

Hi Jeremy,

> I'm wondering if this the right way (statistically speaking) to analyse mixed ploidy datasets in R? Will my estimates of genetic diversity and structure be accurate?

I think the answer is... it depends. Meirmans, Liu, and Tienderen just came out with a review on this topic: https://doi.org/10.1093/jhered/esy006 <https://doi.org/10.1093/jhered/esy006>*. For multivariate analyses (sPCA, PCA, DAPC, etc), you will want to reduce the effect of polyploidy by converting your data to allele frequencies with:

tab(myData, freq = TRUE)

If you have an organism that changes ploidy by life stage, you may need to analyze them separately. Moreover, if you use Bruvo's distance, the choice of model is very important.

> I know there is the POLYSAT package, which seems to have been developed specifically for dealing with mixed ploidy datasets, however, I would rather stick to using adagenet and poppr, as I'm familiar with the functions and the structure of genind and genclone objects.

I know conversion is a PITA, but I would highly recommend using POLYSAT. There are far more tools to deal with polyploidy in that package that can complement any analyses you would perform with poppr or adegenet. A few years ago, I wrote a tiny function to help with conversion from genind to polysat data:

https://gist.github.com/zkamvar/aeaff83b9d126d55aade

In fact, Clarke and Schreier came out with a paper a year ago talking about how to resolve ambiguous ploidy, and it's only available in polysat: http://onlinelibrary.wiley.com/doi/10.1111/1755-0998.12639/abstract


Hope that helps,
Zhian

* The last figure in this paper will change slightly since they used an old version of poppr to calculate Bruvo's distance, which had a silent bug.

-----
Zhian N. Kamvar, Ph. D.
Postdoctoral Researcher (Everhart Lab)
Department of Plant Pathology
University of Nebraska-Lincoln
ORCID: 0000-0003-1458-7108


> 
> Date: Fri, 9 Feb 2018 00:48:39 +0000
> From: JEREMY SAMUEL BENWELL-CLARKE <17197751 at students.latrobe.edu.au>
> To: "adegenet-forum at lists.r-forge.r-project.org"
> 	<adegenet-forum at lists.r-forge.r-project.org>
> Subject: [adegenet-forum] Analysing mixed ploidy datasets
> Message-ID:
> 	<SY3PR01MB15646BCB509043E46D83877FC7F20 at SY3PR01MB1564.ausprd01.prod.outlook.com>
> 
> Content-Type: text/plain; charset="us-ascii"
> 
> Hi everyone,
> 
> I'm trying to analyse a mixed ploidy dataset. The majority of my samples are diploid but there are a few triploids too. To make ploidy even in my raw data matrix I added zeros to all my diploid samples, which makes them triploids. I then use the read.genalex function from the poppr package to read in my data setting ploidy=3. However, I don't want '0' to be recognised as an extra allele and I want the true diploid samples to be separate from the true triploid samples. Therefore, I use the recode_polyploids function from poppr and set newploidy=T. Here is my code:
> 
> genclone<-read.genalex("C:/Users/...", ploidy = 3)
> genclone<-recode_polyploids(genclone, newploidy = T)
> 
> 
> I'm wondering if this the right way (statistically speaking) to analyse mixed ploidy datasets in R? Will my estimates of genetic diversity and structure be accurate?
> I know there is the POLYSAT package, which seems to have been developed specifically for dealing with mixed ploidy datasets, however, I would rather stick to using adagenet and poppr, as I'm familiar with the functions and the structure of genind and genclone objects.
> 
> Any help would be much appreciated!
> 
> Cheers,
> 
> Jeremy
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20180209/7c415c08/attachment-0001.html>
> 
> ------------------------------
> 
> _______________________________________________
> 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
> 
> End of adegenet-forum Digest, Vol 114, Issue 7
> **********************************************

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20180212/5a1b729c/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: Message signed with OpenPGP
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20180212/5a1b729c/attachment.sig>