Bayesian Population Genetic Data Analysis

The notes and files provided here were used in Bayesian parts of the Population Genetic Data Analysis for Natural Populations module in the Summer Institute in Statistical Genetics at the University of Washington, 13-15 June 2007.

A paper providing evidence that athletes use Bayesian learning (instinctively) in experiments designed to simulate tennis appeared in the 15 January 2004 issue of Nature.

Notes

Adobe PDF version

LaTeX source (You'll need the EPS files below to compile the LaTeX source.)

• figure-0.eps

• figure-1.eps

• figure-2.eps

• beta-distribution.eps

• platanthera-allele-freq-dist.eps

• outlier.eps

• outlierdensityplot.eps

Examples

Binomial sampling

Genotype sampling

• Assuming no inbreeding (f = 0)

• Allowing inbreeding (estimating f)

• GC data from African Americans

• Inbreeding with partial dominance (SMFO)

Population structure analysis

• Mosquito data

  • Two alleles per locus (WinBUGS)

  • Multiple alleles per locus (NEXUS)

• Platanthera leucophaea

  • RAPD data (NEXUS)

  • Allozyme data (NEXUS)

Software used or mentioned during the module

WinBUGS

Hickory

R -- 'GNU S', a freely available language and environment for statistical computing and graphics

BOA -- Bayesian Output analysis, routines for convergence diagnostics in R and S-Plus

The Bayesian coin tosser (cointoss.exe) and MCRobot used during the lectures are available from Paul Lewis' software page.

Earlier versions of the notes

If for some reason you are interested in versions of the notes from earlier years, they are also available here.

• 2006

• 2005

• 2004

• 2003

• 2002

Notes and materials provided here are provided under a Creative Commons License.

Kent Holsinger