P-Value for pegas amova

[Kristopher Wojtusik]

Hello,

I am using poppr to run an amova using the pegas method  and everything runs fine, but it does not output any p-values. Is there a way to calculate the p-value, like with the ade4 method?

Pegas outputs a p-value when run on its own, but it cannot handle missing data in the distance matrix, which is why i'm running it through poppr, and also why i'm confused that there is no p-value. 

Thank you for any help you can provide.

-Kris

[Zhian N. Kamvar]

Hello,

Unlike ade4, pegas runs the permutations within the amova function, so you want to use the `nperm` parameter to set the number of permutations. See the documentation: https://grunwaldlab.github.io/poppr/reference/poppr.amova.html

Hope that helps,

Zhian

--
You received this message because you are subscribed to the Google Groups "poppr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to poppr+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/poppr/2c54f2c6-61cd-4ceb-8ccc-31579a916b54n%40googlegroups.com.

[Gabriella Scatà]

Hi, I was just wondering whether you need to calculate adjustment of p-values when using poppr::amova with 1000 permutations, or sicne you are using permutations you don't need to adjust for multiple testing?

And in case you need, is there a way to do so within the poppr::amova?

I would also like to know when I should set "is.squared = TRUE"?

If I am using as a distance matrix Nei's genetic distance, is it correct to set "is.squared = FALSE"? Otherwise, my understanding is that the function will calculate the square root of the genetic distance from the distance matrix provided....correct?

If you instead use Euclidean distance, is that normally calculated as squared and so in that case would we need to set the "is.squared" to TRUE?

Thank you for your kind reply and assistance.

Best

Gabriella