Oral Presentation Society for Molecular Biology and Evolution Conference 2016

A tutorial on how (not) to over-interpret STRUCTURE/ADMIXTURE bar plots       (#228)

Daniel Falush 1 , Lucy van Dorp 2 , Daniel Lawson 3
  1. University of Swansea, Swansea, United Kingdom
  2. University College London, London, United Kingdom
  3. University of Bristol, Bristol, United Kingdom

Genetic clustering algorithms, implemented in popular programs such as STRUCTURE and ADMIXTURE, have been used extensively in the characterisation of individuals and populations based on genetic data. A successful example is reconstruction of the genetic history of African Americans who are a product of recent admixture between highly differentiated populations. Histories can also be reconstructed using the same protocol for scenarios where groups have not experience recent admixture, where recent genetic drift is strong or other scenarios that deviate in some  way from the underlying inference model. Unfortunately, such histories can be highly misleading. We have implemented a "differential palette" visualization of the fineSTRUCTURE coancestry matrix which facilitates easy comparison with STRUCTURE/ADMIXTURE bar plots and assessment of how good a fit the admixture model is for the dataset as a whole and for particular individuals and groups. Combining these complementary analyses with additional methods such as supervised clustering that are designed to test specific hypothesis should allow more robust analysis of recent demographic history based on genetic data.