[R-sig-phylo] Determining phylogenetic signal in non-ultrametric trees

Nina Hobbhahn n.hobbhahn at gmail.com
Wed Feb 15 12:13:26 CET 2012


Dear fellow list users,

when using fitDiscrete to determine the strength of phylogenetic signal of a discrete trait over a set of non-ultrametric phylograms, I get the warning "Warning: some tree transformations in GEIGER might not be sensible for nonultrametric trees.". Research into this topic has led me to Carl Boettiger's online notebook, where he states "No, the transformation is only meaningful it the trees are ultrametric. The models describe rates of change through time, and thus assume branch lengths are proportional to time, not to number of substitutions. This is true for most models in comparative methods." (http://openwetware.org/wiki/User:Carl_Boettiger/Notebook/Comparative_Phylogenetics/2010/10/12). However, Litsios and Salamin's upcoming paper in Systematic Biology (DOI:10.1093/sysbio/syr124) indicates that the authors have calculated both Pagel's lambda and Blomberg's K over non-ultrametric phylograms, and compared the results to lambda and K obtained from ultrametric chronograms. I would be grateful for some opinions on the following two questions:

1. Is the lambda tree-transformation in fitDiscrete sensitive to non-ultrametric trees?
2. Is determination of phylogenetic-signal strength from (non-ultrametric) phylograms a valid approach?

Thank you very much,

Nina


Dr. Nina Hobbhahn
Post-doctoral fellow
Lab of Prof. S. D. Johnson
School of Life Sciences
University of KwaZulu-Natal
Private Bag X01
Scottsville, Pietermaritzburg, 3201
South Africa



More information about the R-sig-phylo mailing list