25th Annual Student Research Conference
Separating the Effects of Natural Selection and Genetic Drift Using a Modification of Tajima's D Statistic
Nathan D. Kopp* and Colin D. Teberg
Dr. Anton Weisstein and Mr. Pamela J. Ryan, Faculty Mentors
Many statistical tests have been devised to infer selective processes and demographic structure from the patterns of nucleotide variation within a population. However, interpretation of the results is complicated by the fact that a wide range of population-level processes can yield similar patterns within the data. Here we propose and evaluate a partitioning approach that separately analyzes synonymous and nonsynonymous variation. Using modified versions of Tajimas D statistic, we demonstrate that this approach successfully distinguishes data sets simulated under a neutral evolutionary model from those simulated under purifying selection, diversifying selection, and population subdivision. These last three models are also rejected in an analysis of actual genetic sequences from the V3 loop of HIV. However, population bottlenecks could not be resolved from the neutral model. Power analyses confirm that data partitioning approaches have promise but require further refinements to increase their statistical power.
Keywords: Mutation, evolution, partitioning, HIV, statisitcal power, selective processes, demographic structure, Tajima's D
Topic(s):Mathematical Biology
Biology
Statistics
Presentation Type: Oral Paper
Session: 210-3
Location: VH 1228
Time: 10:00