Thanks for the input. I'll look into those publications and the Rphylopars
package.
Actually, all of my data already has uncertainty. I've been getting mean
values from published studies and then weighting these by the inverse
variance (as is common in meta-analyses). So I have both uncertainty in
measurements (from multiple measures within a single study) AND multiple
measurements (across different studies). I haven't combined estimates
across studies because I think one of the most important predictors is the
study ID/author.
Philip Shirk
PhD Candidate
Osenberg Lab
Department of Biology
University of Florida
616 Bartram
P.O. Box 118525
Gainesville FL, 32611-8525
On Fri, Jan 27, 2017 at 12:08 PM, Christopher Hamilton <
cah0...@tigermail.auburn.edu> wrote:
> If you have multiple measurements per species (multiple individuals of a
> species, to represent a single tip), there is a program Rphylopars that is
> now available, that accounts for within-species variation.
>
> Though maybe it’s not appropriate for your data, as Liam was saying.
>
> I like it, but I would love to hear from those individuals here that are
> much more qualified than I am…
>
> Chris
>
>
> > On Jan 27, 2017, at 12:00 PM, Liam J. Revell
> wrote:
> >
> > Hi Philip.
> >
> > Do not create very short tips. This will actually lead to very high
> weight being given to any difference between these two observations! Take
> the average of the different observations. Using the methods of Felsenstein
> (2008) or Ives & Garland (2007) it is also possible to take into account
> within-species sampling error, though this is not precisely what you have
> here (you do not know how uncertain each measure is, nor how uncertain are
> your estimates for other species), nor is this implemented in R.
> >
> > All the best, Liam
> >
> > Liam J. Revell, Associate Professor of Biology
> > University of Massachusetts Boston
> > web: http://faculty.umb.edu/liam.revell/
> > email: liam.rev...@umb.edu
> > blog: http://blog.phytools.org
> >
> > On 1/27/2017 11:45 AM, Philip Shirk wrote:
> >> I know this is a repeat question (
> >> http://www.mail-archive.com/r-sig-phylo@r-project.org/msg04424.html)
> but I
> >> didn't see any reponses/answers to the first posting, so I thought I'd
> ask
> >> again:
> >>
> >> I have a pre-published tree for my species of interest that includes all
> >> the species in my dataset. However, my dataset contains duplicate data
> for
> >> a few species (different estimates from different authors). What is the
> >> best way to incorporate all the data into a PGLS analysis? When I try to
> >> run the analysis with gls() I get the error:
> >> "Error in Initialize.corPhyl(X[[i]], ...) : number of observations and
> >> number of tips in the tree are not equal."
> >>
> >> I imagine that I could:
> >> 1. toss out duplicates
> >> 2. average/combine duplicates
> >> 3. create very short branches for duplicate species and treat them as
> very
> >> closely-related but separate species (but I don't know how to actually
> do
> >> this.)
> >> 4. ???
> >>
> >> Thanks,
> >> Philip Shirk
> >>
> >> [[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-ph...@r-project.org/
> >>
> >
> > ___
> > 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-ph...@r-project.org/
>
>
[[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/
