I think the way to go is to treat it as separate characters: using the
zero-length branches assumes that the different seasonal morphs are
separate populations that evolved different phenotypes
instantaneously... the Markov model is going to consider that very
strange, and think transition rates are extremely high.
One thing to consider is that there are different ways of breaking a
complex trait into multiple discrete traits, and you may want to
consider the potential that given how you break it up, the rate or
symmetry of evolutionary transitions for one character may depend on
the other character. How to handle it probably depends on your
colleague's specific question.
Cheers,
-Dave
On Tue, May 23, 2017 at 9:48 AM, Jacob Berv
wrote:
> Dear R-sig-phylo,
>
> I was wondering - is anyone aware of methods or models that can deal with
> traits that are have evolved seasonal discrete plasticity in some lineages,
> whereas in other lineages such seasonality has not evolved (and so traits
> evolve as a discrete character?). I’m helping a colleague who wants to
> estimate transition rates in a group of butterflies for particular color
> patterns.
>
> I have been thinking that a workaround might be to code zero length terminal
> branches for lineages which have seasonal plasticity so that some of that
> information can be incorporated into ML reconstructions. Or to simply code
> two sets of characters for with ’season’ assumed to be homologous. Any ideas?
>
> best,
> Jake Berv
> ___
> 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/
--
David W. Bapst, PhD
https://github.com/dwbapst/paleotree
___
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/
