A Closer Look at Models Developed for Landscape Genetics

• Activity Number: 250
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Kezia Manlove*+ and Megan Higgs
• Companies: Montana State University and Montana State University
• Keywords: Bayesian statistics ; Spatial statistics ; Population genetics

Abstract

Genetic data contain a wealth of untapped information valuable to ecologists. One area of growing ecological interest is the use of spatially point-referenced genetic data to locate individuals and populations on a landscape, in an effort to delineate among separate breeding groups. We investigate a popular Bayesian model for inferring spatial population structure and ranges based on point-referenced genetic data. Through a series of simulations, we assess the model's capacity to correctly assign individuals to populations, examine the closeness of the model-inferred genetic structure to parametric structure, and observe the accuracy of maps of the model-defined populations. Additionally, we compare the Bayesian model's ability to address these questions to a commonly implemented clustering algorithm, and discuss limitations on the model's scope of inference.