Hi Emmanuel,
Thank you so much for taking the pain to check this. Looking at
my vector of frequencies (Pfreq[,2]) it is pretty clear that it is
hard to believe that the trait variance conforms to BM.
All the best
Juan
El 18/05/2017 a las 10:00, Emmanuel
Paradis escribió:
Hi Juan,
This warning seems related to the fact that the trait did not
evolve according to the specified model (i.e., Brownian motion). I
simulated some data with:
tr <- rphylo(50, .1, 0)
x <- rTraitCont(tr)
ace(x, tr)$sigma did not return NA out of 1000 replications. I
changed the second line to:
x <- rnorm(50)
and NAs were returned in 18 cases.
In your case, since you simulated the data, it could be that their
distribution does not conform well to the Brownian motion model.
Maybe you could transform them or discretize them and use a
discrete trait model (i.e., Markovian model, type = "d" in ace).
HTH
Best,
Emmanuel
Le 11/05/2017 à 15:54, Juan Antonio Balbuena a écrit :
Hi all
I am playing around with synthetic cophylogenetic data. In one
example, which represents a very extreme case of high agreement
between the host and parasite phylogenies (the data was
generated with 51 cospeciation, 2 sorting and 1 duplication
events), I get a warning when using ace() to label the
cophylogenetic nodes of the parasite tree:
fit.P<-ace(Pfreq[,2],obj$trees[[2]])
Warning message:
In sqrt(1/out$hessian) : NaNs produced
Where Pfreq[,2] is a vector of frequencies:
> Pfreq[,2]
[1] 100.0 100.0 69.70874 100.0 100.0
100.0 100.0 100.0
[9] 100.0 100.0 100.0 37.86408 38.44660
62.13592 61.55340 100.0
[17] 100.0 100.0 100.0 100.0 100.0 100.0
100.0 100.0
[25] 100.0 100.0 100.0 100.0 62.13592 100.0
100.0 100.0
[33] 100.0 100.0 100.0 100.0 100.0 100.0
100.0 100.0
[41] 100.0 100.0 68.15534 100.0 100.0 100.0
100.0 100.0
[49] 100.0 100.0 100.0 100.0 100.0
The following works fine
nodelabels.cophylo(pie=fit.P$ace/100,
piecol=c("green","red"),cex=0.5, which="right")
but I have noticed that there is a NA in fit.P$sigma2. It would
be nice to know whether this warning is important or not.
Thank you very much for your attention.
Juan A. Balbuena
--
Dr. Juan A. Balbuena
Cavanilles Institute of Biodiversity and Evolutionary Biology
University of Valencia http://www.uv.es/~balbuena
P.O. Box 22085 http://www.uv.es/cophylpaco
46071 Valencia, Spain
e-mail: j.a.balbu...@uv.es tel.
+34 963 543 658 fax +34 963 543 733
*NOTE!*For shipments by EXPRESS COURIER use the following street
address:
C/ Catedrático José Beltrán 2, 46980 Paterna (Valencia), Spain.
Pour nous remonter une erreur de filtrage, veuillez vous rendre
ici
___
R-sig-phylo mailing list - R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
Searchable archive at
http://www.mail-archive.com/r-sig-phylo@r-project.org/
Pour nous remonter une erreur de filtrage, veuillez vous rendre
ici : http://f.security-mail.net/301EU0Nrv8J
--
Dr.
Juan
A. Balbuena
Cavanilles Institute of Biodiversity and Evolutionary
Biology
University of
Valencia
http://www.uv.es/~balbuena
P.O. Box
22085
