Hi everyone:
I'm trying to do a phylogenetic ANOVA analysis, but i'm stuck with both
possibilities: phylANOVA from phytools package, and aov.phylo from geiger
package.
Here is what i have
Input data:
CSV file with columns:
Spp (categorical, for taxonomic species,, 29 species)
Fam (categorical, for taxonomic families)
IF (continuous, length ratio data ranging from 1 to 3 approx, 91 cases)
Phylogenetic tree. (Newick format, with distances, 29 species). I'm sure
that all species are both in phylo tree and in data.
My problems are:
phylANOVA didn't calculate post-hoc tests. All i got was an F and p, but
all results from post-hocs are NA.
aov.phylo won't stop saying:
'formula' must be of the form 'dat~group', where 'group' is a named factor
> vector and 'dat' is a data matrix or named vector"
When were made this way:
dat <- as.vector(data$IF)
names(dat)<-names(data$Spp)
group <- as.vector(data$Spp)
names(group)<-names(data$Spp)
I've googled and tried several ways to name the damned vectors, but it
keeps saying that. Like: names(OBJECT)<-c("name1", "name2",....)
Eventually, i could run it, i don't know how (and that bothers me). BUT,
something bad happened:
The following tips were not found in 'phy' and were dropped from 'data':
>
And viceversa for 'data' and 'phy'. So, analysis is restricted to F and p
values. When i try to make the juicy post-hoc tests (even when i don't know
if it will work), it fails saying that Tukey HSD is not appliable to lm
object.
Hope you understood me and my english, and can help me!
Thank you very much in advance
Guido, from Mar del Plata
--
Lic. Guido Buezas
Laboratorio de Morfología Funcional y Comportamiento
IIMyC - UNMdP / CONICET
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
