Abstract Detail

Phylogenetic Comparative Methods and Approaches in Plant Science

Beaulieu, Jeremy M. [1].

Exploring heterogeneity in binary character evolution.

Biologists now have the capability of building large phylogenetic trees consisting of tens of thousands of species, from which important comparative questions can be addressed. However, to the extent that biologists have applied these large trees to comparative data, it is clear that current methods, such as those that deal with the evolution of binary morphological characters, can make unrealistic assumptions about how these characters are modeled. As phylogenies increase both in size and scope it is likely that the lability of a binary character will differ significantly among lineages. I will describe how a new generalized model, which I refer to as the "hidden rates model" (HRM), can be used to identify different rates of evolution in a discrete binary character along different branches of a phylogeny. The HRM is part of a class of models that are more broadly known as Hidden Markov models because it presupposes that unobserved "hidden" rate classes underlie each observed state, and that each rate class represents potentially different transition rates to and from these observed states. As I will discuss, the recognition and accommodation of this heterogeneity can provide a robust picture of binary character evolution across very large, very old, and very widespread flowering plant clades. I will also demonstrate further extensions of these models, which includes a novel measure of evolutionary potential of contemporary species as well as a model that allows for a more refined understanding of how particular character states influence the diversification process.

1 - NIMBioS, 1122 Volunteer Blvd, Ste. 106, Knoxville, TN, 37996, USA

Keywords:
Hidden rates model
Character evolution
flowering plants.

Presentation Type: Symposium or Colloquium Presentation
Session: SY01
Location: Evergreen/Grove
Date: Monday, July 28th, 2014
Time: 8:15 AM
Number: SY01002
Abstract ID:641
Candidate for Awards:None