Estimating Selection Coefficients in Spatially Structured Populations from Time Series Data of Allele Frequencies

• DOI: 10.1534/genetics.112.147611
• Publication Date: 2013-01-11T02:06:58Z
• Authors: Iain Mathieson; Gil McVean
• Keywords: 0301 basic medicine; Time Factors; Models, Genetic; Pigmentation; Population; Investigations; Moths; Evolution, Molecular; 03 medical and health sciences; Gene Frequency; Animals; Animal Migration; Selection, Genetic

Inferring the nature and magnitude of selection is an important problem in many biological contexts. Typically when estimating a coefficient for allele, it assumed that samples are drawn from panmictic population acts uniformly across population. However, these assumptions rarely satisfied. Natural populations almost always structured, selective pressures likely to act differentially. Inference about ought therefore take account structure. We do this by considering evolution simple lattice model spatial develop hidden Markov based maximum-likelihood approach single time series data allele frequencies. then approximate extension structured case provide joint estimate migration rate spatially varying coefficients. illustrate our method using classical sets moth pigmentation morph frequencies, but has wide applications settings ranging ecology human evolution.