Department of Mathematics Colloquium

University of Idaho

Spring 2012

Thursday,  April 5, 3:30-4:20 pm, room TLC 032

Refreshments in Brink 305 at 3:00 pm

The Dimensionality of Evolution


Paul Hohenlohe

Institute for Bioinformatics and Evolutionary Studies

Departments of Biological Sciences, Statistics

University of Idaho


The dynamic behavior of evolving biological systems involves a vast number of factors, including environmental variables, complex interactions among species, and numerous organismal traits such as morphology, physiology, and behavior.  This implies a highly multidimensional space in which evolution can operate.  However, tight relationships among factors, such as strong genetic correlations among traits, can constrain the trajectories of evolution to a much smaller subspace.  I will discuss a novel technique for estimating the number of independent axes along which evolution can operate – the dimensionality of evolution – based on empirical data in the form of interaction matrices.  I will illustrate the approach for estimating the dimensionality of reproductive isolation, the critical step in the formation of new species.  Based on datasets from several animal groups, this dimensionality appears to have a strong upper limit, and mapping of independent axes can provide new insight into the underlying biological processes.  The method is easily generalizable to a wide range of biological situations, and the type of empirical data required to draw inferences is commonly gathered in ecological and evolutionary studies.