E&E Seminar: “New tools for the analysis and visualization of geographic structure in population genetic data” by John Novembre, University of Chicago
February 11, 2020
12:00 PM - 1:00 PM
Population structure is a fundamental feature of genetic variation that is crucial to address when carrying out studies in evolutionary biology, conservation genetics, and trait mapping. However, most analytical methods to represent population structure do not incorporate geography directly. Typically, geography must be considered post hoc alongside a visual summary of genetic data. Recently, we developed a spatially explicit method that estimates “effective migration surfaces” to visualize how genetic diversity is geographically structured (the EEMS method). Here, we share insights from applications of the EEMS method using examples from multiple species, leveraging both published studies and novel applications. In particular, using a novel analysis of over 8,000 widespread human samples we find surfaces that are “rugged”, which indicates the relationship between genetic and geographic distance is heterogeneous and subtly distorted as a rule. We also highlight possible pitfalls of applying the method and share novel lines of development of the EEMS method, including an extension that analyzes long-shared haplotype tracts, which can be used to assess population structure on different time-scales. I will also show a complementary technique for making clear the underlying allele frequency patterns that are common among large collections of human single nucleotide variants (SNVs). These results provide visualizations of human genetic diversity that reveal local patterns of differentiation in detail. The fine-scale population structure depicted here is relevant for understanding complex processes of human population history and the tools have application to a diverse range of species.
Faculty Host: Boris Igic
Date posted
Nov 20, 2019
Date updated
Feb 3, 2020