Abstract Detail

Hybrids and Hybridization

Landis, Henry [1], Eaton, Deren [2].

A probabilistic model for inferring hybridization across large taxonomic groups, with applications to Quercus.

To understand hybridization, it is necessary to understand the process of how species become reproductively isolated. Lineages evolve reproductive isolation (RI) at different rates, but the underlying assumption that RI evolution rates correlate with speciation rates has rarely been tested. The genus Quercus (oaks) has a deep and challenging legacy of introgression within its ranks and represents a promising avenue for investigating hybridization patterns. Here we describe a Bayesian hierarchical linear modeling approach for inferring pairwise RI as a function of the genetic distance between all species in a phylogeny, based on sparse observations of hybridization between many species pairs. Using simulations we show that our probabilistic model returns accurate posterior estimates of RI rates even for very sparse datasets. When applied to real hybrid observations in Quercus, we explored several different ways of encoding observations, including the inference of hybrid absence from a combination of geographic overlap, abundant observations, and a lack of hybrid descriptions in the literature. Our results indicate that posterior estimates of RI rates vary among major oak sections, and demonstrate the potential to characterize patterns of isolation, speciation and hybridization across a wide range of taxa from limited empirical data.

1 - Columbia University, Ecology, Evolution and Environmental Biology, 1200 Amsterdam Avenue, 10th Floor Schermerhorn EXT, New York, NY, 10027, United States
2 - Columbia University, Ecology, Evolution, And Environmental Biology, 1200 Amsterdam Ave. , Schermerhorn Ext. Office 1007, New York, NY, 10027, United States

Keywords:
hybridization
modeling
Quercus.

Presentation Type: Oral Paper
Session: HH2, Hybrids and Hybridization II
Location: /
Date: Wednesday, July 21st, 2021
Time: 1:00 PM(EDT)
Number: HH2003
Abstract ID:842
Candidate for Awards:Margaret Menzel Award