[adegenet-forum] Scripting adegenet with whole-genome data

[Rachel Turba]

Hi Brian

Yes, I think I will have to try that. I am not sure either if it will work but in the tutorial there was a step that looked at SNPs contributing to the variation in the set. I just thought it would be worth trying as an alternative tool alongside genome wide Fst.

Thank you so much for the advice and help.

Cheers,
Rachel

________________________________
From: brian knaus <briank.lists at gmail.com>
Sent: Wednesday, March 6, 2019 1:27:02 PM
To: Rachel Turba
Cc: adegenet-forum at lists.r-forge.r-project.org
Subject: Re: [adegenet-forum] Scripting adegenet with whole-genome data

Hi Rachel,

I believe that the statement for why DAPC can not be scripted is because the steps to determine the number of principal components and discriminant functions are interactive. If you know this information you can send the DAPC step to a server without a monitor. Not sure DAPC is going to help you find SNPs associated with phenotypes though.

Good luck!
Brian

On Wed, Mar 6, 2019 at 11:14 AM Rachel Turba <rturba at outlook.com<mailto:rturba at outlook.com>> wrote:
Hi Brian,

Thank you for the help. I was avoiding to subset my data because I am interested in investigating SNPs that could be related to a change in morph between these populations.

Cheers,
Rachel

________________________________
From: brian knaus <briank.lists at gmail.com<mailto:briank.lists at gmail.com>>
Sent: Tuesday, March 5, 2019 2:05:03 PM
To: Rachel Turba
Cc: adegenet-forum at lists.r-forge.r-project.org<mailto:adegenet-forum at lists.r-forge.r-project.org>
Subject: Re: [adegenet-forum] Scripting adegenet with whole-genome data

Hi Rachel,

Demographic patterns are generally thought to occur throughout the genome. In which case you shouldn't need the entire genome but can use a representative subset of your variants. We've demonstrated how to accomplish this at the below link.

https://grunwaldlab.github.io/Population_Genetics_in_R/gbs_analysis.html#subsetting-a-vcfr-object-to-200-random-variants

Good luck!
Brian

On Tue, Mar 5, 2019 at 11:24 AM Rachel Turba <rturba at outlook.com<mailto:rturba at outlook.com>> wrote:
Hi all,

I have researched this on the forum but was not successful in finding help, so I apologize in advance if this has been addressed already.

In this tutorial (https://grunwaldlab.github.io/Population_Genetics_in_R/DAPC.html) it says that adegenet cannot be scripted a priori, but my data has 6Gb and takes a very long time to generate the DAPC object. So I have to submit it as a script to the university cluster. I get the plot of variance explained by PC but then everything halts because I do not provide the number of PCs to be retained for continuing the analysis. Is there a way around this? How does anyone handle large datasets?

Thank you in advance,
Rachel
_______________________________________________
adegenet-forum mailing list
adegenet-forum at lists.r-forge.r-project.org<mailto: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<mailto:adegenet-forum at lists.r-forge.r-project.org>
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/adegenet-forum
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20190306/1ba5a835/attachment-0001.html>