Skip to content
Prev 324666 / 398503 Next

Mixed effects model with a phylogenetic tree/ distance matrix as a random effect

Bert Gunter <gunter.berton <at> gene.com> writes:
Seconded.  Alternatively you could try the r-sig-phylo at r-project.org
mailing list, although I think I would try R-s-m-m first.
This is not going to change in the near future, although
there is a 'pedigreemm' package built on lme4 that 
might do what you want.

  You might be able to use MASS::glmmPQL in conjunction with
the corStruct structures from nlme (for classical geostatistical
correlation models); ape (for phylogenetic models); or ramps (other
choices).

  INLA might ??? allow tree correlation structures:
http://arxiv.org/abs/1210.4908

  Ives and Garland have some MATLAB code for phylogenetic 
logistic regression ...