[adegenet-forum] confusing p value in mantel test

[Vojtěch Zeisek]

Hello

Dne St 18. června 2014 11:45:00, Kelly Bennett napsal(a):
> Hello,
> 
> I have run a mantel test with the following code
> 
> dna <- read.dna(file = "dna_manteltest.fasta", format = "fasta")
> dna.dists <- dist(dna, method = "euclidean")

Why do You use function dist() and not dist.dna() (package APE) having various 
mutations models? IMHO, Euclidean distance is not the best for nucleotide 
data, I'd use it for fragmentation data, but not here.

> as.matrix(dna.dists)[1:5, 1:5]
> geo <- read.csv(file = "geo_matrix.csv")
> geo[1:2, 1:2]
> geo.dists <- dist(geo, method = "euclidean")
> as.matrix(geo.dists)[1:5, 1:5]
> mantelresult<-mantel.rtest(dna.dists, geo.dists, nrepet = 9999)
> cor.test(geo.dists, dna.dists)
> plot(mantelresult <- mantel.rtest(dna.dists, geo.dists), main = "Mantel's
> test") mantelresult
> 
> From my plot it looks like there should be isolation by distance and a
> correlation test shows a significant association but my p value for the
> Monte Carlo test = 1
> 
> Does anyone have any ideas about this contradiction? I have attached the
> plot to this email

Well, it will produce some result every time You give it some data, even if 
they are wrongly used. Right now it might be the case.

> Thank you very much,
> 
> Kelly

Sincerely,
Vojtěch

-- 
Vojtěch Zeisek
http://trapa.cz/en/

Department of Botany, Faculty of Science
Charles University in Prague
Benátská 2, Prague, 12801, CZ
http://botany.natur.cuni.cz/en/

Institute of Botany, Academy of Science
Zámek 1, Průhonice, 25243, CZ
http://www.ibot.cas.cz/en/

Czech Republic
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.r-forge.r-project.org/pipermail/adegenet-forum/attachments/20140623/f24bde66/attachment.sig>