----- Le 17 Mai 21, à 13:41, Marguerite Butler <mbutler...@gmail.com> a écrit :
> Thank you very much for the reply Emmanuel!
> OK, yes I just tried and Iʻm surprised that this works:
>> dat <- data.frame(Species = c("Homo", "Pongo", "Macaca", "Ateles",
>> "Galago"), X
> > = X, Y = Y)
> > cbm <- corBrownian(1, tree, form = ~Species)
> > gls(Y ~ X, dat, correlation = cbm)
> How does corBrownian know about Species, when it is only defined as an element
> of dat, and not known in the global environment?
corBrownian() doesn't know where all these variables are located. Indeed, if
you print 'cbm':
R> cbm
Uninitialized correlation structure of class corBrownian
It's gls() that looks for them. When you think about it, using "data = ..." is
really a good way to be sure that all variables are ordered correctly whether
they are involved as predictors (in 'model = '), in the correlation structure,
or in the variance function ('weights = ').
Best,
Emmanuem
> Anyway, thank you!
> Marguerite
> On Sun, May 16, 2021 at 7:03 PM Emmanuel Paradis < [
> mailto:emmanuel.para...@ird.fr | emmanuel.para...@ird.fr ] > wrote:
>> Hi Marguerite,
>> The issue is about nlme::gls. The help page ?gls says:
>> data: an optional data frame containing the variables named in
>> ‘model’, ‘correlation’, ‘weights’, and ‘subset’. By default
>> the variables are taken from the environment from which ‘gls’
>> is called.
>> So you make it work by removing "dat$" in the call to corBrownian():
>> cbm <- corBrownian(1, tree, form = ~Species)
>> Best,
>> Emmanuel
>> ----- Le 17 Mai 21, à 6:22, Marguerite Butler [ mailto:mbutler...@gmail.com |
>> mbutler...@gmail.com ] a écrit :
>> > Aloha all,
>> > Does anyone know the answer to this question - how does ape pass the
>> > species name information into gls? Seems to be a global environment issue
>> > when using apeʻs corBrownian and nlmeʻs gls? and a very specific form is
>> > required, but why?
>> > Modified from ape help page for CorClasses:
>> > ---
>> > library(nlme)
>> > library(ape)
>> > tree <- read.tree(text =
>> > "((((Homo:0.21,Pongo:0.21):0.28,Macaca:0.49):0.13,Ateles:0.62):0.38,Galago:1.00);")
>> > X <- c(4.09434, 3.61092, 2.37024, 2.02815, -1.46968)
>> > Y <- c(4.74493, 3.33220, 3.36730, 2.89037, 2.30259)
>> > dat <- data.frame(Species = c("Homo", "Pongo", "Macaca", "Ateles",
>> > "Galago"), X = X, Y = Y)
>> > gls(Y ~ X, dat, correlation=corBrownian(1, tree.primates, form = ~Species))
>> > ---
>> > The above works just fine. However, if we try to calculate corBrownian
>> > separately, gls cannot understand the corBrownian object:
>> > ---
>> >> cbm <- corBrownian(1, tree, form = ~dat$Species) # calculate correlation
>> >> separately
>> >> gls(Y ~ X, dat, correlation = cbm) ## does not work, invalid type
>> >> (list) for dat
>> > Error in model.frame.default(formula = ~dat + Species + Y + X, data =
>> > list( :
>> > invalid type (list) for variable 'dat'
>> > ---
>> > If we put the species information into a vector in the global
>> > environment it works. Why canʻt we use dat$Species?
>> > ---
>> > spp <- dat$Species # save species as vector in global environment
>> > cbm <- corBrownian(1, tree, form = ~spp) # calculate correlation separately
>> > gls(Y ~ X, dat, correlation = cbm) ## works!
>> > Thanks,
>> > Marguerite
>>> On Fri, May 14, 2021 at 10:21 PM Marguerite Butler < [
>> > mailto:mbutler...@gmail.com | mbutler...@gmail.com ] >
>> > wrote:
>> >> Aloha Oliver,
>> >> From the cor.Brownian help page, the explanation for the form argument is
>> >> this:
>> >> a one sided formula of the form ~ t, or ~ t | g, specifying the taxa
>> >> covariate t and, optionally, a grouping factor g. A covariate for this
>> >> correlation structure must be character valued, with entries matching the
>> >> tip labels in the phylogenetic tree. When a grouping factor is present in
>> >> form, the correlation structure is assumed to apply only to observations
>> >> within the same grouping level; observations with different grouping
>> >> levels
>> >> are assumed to be uncorrelated. Defaults to ~ 1, which corresponds to
>> >> using
>> >> the order of the observations in the data as a covariate, and no groups.
>> >> This means that "t" should contain the species names or tip labels,
>> >> spelled exactly as they are in your phylogeny. So for example if your data
>> >> SteninaeData
>> >> has a column for species names called "species" then you would specify:
>> >> cor.BM <- corBrownian(phy=SteninaeTree, form = ~dat$species)
>> >> and then proceed with your regression model as above.
>> >> You can also follow the example on the corClasses page:
>>>> [ http://127.0.0.1:27386/library/ape/html/corClasses.html |
>> >> http://127.0.0.1:27386/library/ape/html/corClasses.html ]
>> >> Slightly modified here:
>> >> ---
>> >> library(nlme)
>> >> library(ape)
>> >> tree <-
>> >> read.tree(text="((((Homo:0.21,Pongo:0.21):0.28,Macaca:0.49):0.13,Ateles:0.62):0.38,Galago:1.00);"
>> >> X <- c(4.09434, 3.61092, 2.37024, 2.02815, -1.46968) # a vector
>> >> Y <- c(4.74493, 3.33220, 3.36730, 2.89037, 2.30259) # a vector
>> >> Species <- c("Homo", "Pongo", "Macaca", "Ateles", "Galago") # a vector
>> >> dat <- data.frame(Species = Species, X = X, Y = Y) # make a dataframe
>> >> with Species, X, Y, can also read in .csv
>> >> cbm <- corBrownian(1, tree, form = ~dat$Species)
>> >> m1 <- gls(Y ~ X, dat, correlation=cbm)
>> >> m1
>> >> ---
>> >> hth,
>> >> Marguerite
>>>> On Fri, May 14, 2021 at 9:46 PM Oliver Betz < [
>> >> mailto:oliver.b...@uni-tuebingen.de | oliver.b...@uni-tuebingen.de ] >
>> >> wrote:
>> >>> Dear all:
>> >>> I am currently facing the following problem:
>> >>> Unsing ape, nlme por phylolm for PGLS I was wondering how to sort my
>> >>> trait data (the ones in my csv file) according to the tree, so that
>> >>> both are in the same order. I learned how to check whether the data &
>> >>> the tree coincide in both number of species and names, but I do not
>> >>> nderstand how to perform the mentioned sorting task.
>> >>> Once I am running the PGLS commands such as
>> >>> cor.BM <- corBrownian(phy=SteninaeTree)
>> >>> pgls <- gls(LOG_rel_Attachment_force_smooth ~
>> >>> LOG_rel_Anzahl_Hafthaare_4.Vordertarsus, correlation = cor.BM, data =
>> >>> SteninaeData, method = "ML")"
>> >>> I regularly receive the follwing warning note:
>> >>> "In Initialize.corPhyl(X[[i]], ...) :
>> >>> No covariate specified, species will be taken as ordered in the data
>> >>> frame. To avoid this message, specify a covariate containing the
>> >>> species names with the 'form' argument."
>> >>> Since this is just a warning message, I usually was going on with the
>> >>> "summary(pgls)" command, where I was getting my results.
>> >>> Anyhow, I was wondering whether it is OK to just ignore this warning
>> >>> or whether it means that I will have to sort the sequence of the
>> >>> sepcies trait data in my data file (csv) exactly according to the
>> >>> sequence of the species as they occur at the tips of my tree file.
>> >>> Since the spelling of the species names is exactly the same in both
>> >>> the tree file and the data file, I thought that such sorting is
>> >>> automatically done by the algorithm?
>> >>> I will be glad if you have any hint how ro resove this.
>> >>> My best wishes,
>> >>> Oliver Betz
>> >>> (University of Tübingen, Gernmany)
>> >>> _______________________________________________
>>>>> R-sig-phylo mailing list - [ mailto:R-sig-phylo@r-project.org |
>> >>> R-sig-phylo@r-project.org ]
>>>>> [ https://stat.ethz.ch/mailman/listinfo/r-sig-phylo |
>> >>> https://stat.ethz.ch/mailman/listinfo/r-sig-phylo ]
>> >>> Searchable archive at
>>>>> [ http://www.mail-archive.com/r-sig-phylo@r-project.org/ |
>> >>> http://www.mail-archive.com/r-sig-phylo@r-project.org/ ]
>> >> --
>> >> ____________________________________________
>> >> Marguerite A. Butler
>> >> Professor
>> >> Department of Biology
>> >> 2538 McCarthy Mall, Edmondson Hall 216
>> >> Honolulu, HI 96822
>> >> Office: 808-956-4713
>> >> Dept: 808-956-8617
>> >> Lab: 808-956-5867
>> >> FAX: 808-956-4745
>> >> [ http://butlerlab.org/ | http://butlerlab.org ]
>>>> [ http://manoa.hawaii.edu/biology/people/marguerite-butler |
>> >> http://manoa.hawaii.edu/biology/people/marguerite-butler ]
>> >> [ http://www2.hawaii.edu/~mbutler | http://www2.hawaii.edu/~mbutler ]
>> > --
>> > ____________________________________________
>> > Marguerite A. Butler
>> > Professor
>> > Department of Biology
>> > 2538 McCarthy Mall, Edmondson Hall 216
>> > Honolulu, HI 96822
>> > Office: 808-956-4713
>> > Dept: 808-956-8617
>> > Lab: 808-956-5867
>> > FAX: 808-956-4745
>> > [ http://butlerlab.org/ | http://butlerlab.org ]
>>> [ http://manoa.hawaii.edu/biology/people/marguerite-butler |
>> > http://manoa.hawaii.edu/biology/people/marguerite-butler ]
>> > [ http://www2.hawaii.edu/~mbutler | http://www2.hawaii.edu/~mbutler ]
>> > [[alternative HTML version deleted]]
>> > _______________________________________________
>>> R-sig-phylo mailing list - [ mailto:R-sig-phylo@r-project.org |
>> > R-sig-phylo@r-project.org ]
>>> [ https://stat.ethz.ch/mailman/listinfo/r-sig-phylo |
>> > https://stat.ethz.ch/mailman/listinfo/r-sig-phylo ]
>>> Searchable archive at [
>>> http://www.mail-archive.com/r-sig-phylo@r-project.org/ |
>> > http://www.mail-archive.com/r-sig-phylo@r-project.org/ ]
> --
> ____________________________________________
> Marguerite A. Butler
> Professor
> Department of Biology
> 2538 McCarthy Mall, Edmondson Hall 216
> Honolulu, HI 96822
> Office: 808-956-4713
> Dept: 808-956-8617
> Lab: 808-956-5867
> FAX: 808-956-4745
> [ http://butlerlab.org/ | http://butlerlab.org ]
> [ http://manoa.hawaii.edu/biology/people/marguerite-butler |
> http://manoa.hawaii.edu/biology/people/marguerite-butler ]
> [ http://www2.hawaii.edu/~mbutler | http://www2.hawaii.edu/~mbutler ]
[[alternative HTML version deleted]]
_______________________________________________
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/
