UC Berkeley

Motivation

Structure methods are highly used population genetic methods for classifying individuals in a sample fractionally into discrete ancestry components.

Contribution

We introduce a new optimization algorithm of the classical Structure model in a maximum likelihood framework. Using analyses of real data we show that the new optimization algorithm finds higher likelihood values than the state-of-the-art method in the same computational time. We also present a new method for estimating population trees from ancestry components using a Gaussian approximation. Using coalescence simulations modeling populations evolving in a tree-like fashion, we explore the adequacy of the Structure model and the Gaussian assumption for identifying ancestry components correctly and for inferring the correct tree. In most cases, ancestry components are inferred correctly, although sample sizes and times since admixture can influence the inferences. Similarly, the popular Gaussian approximation tends to perform poorly when branch lengths are long, although the tree topology is correctly inferred in all scenarios explored. The new methods are implemented together with appropriate visualization tools in the computer package Ohana.

Availability

Ohana is publicly available at https://github.com/jade-cheng/ohana . Besides its source code and installation instructions, we also provide example workflows in the project wiki site.

Contact

jade.cheng@birc.au.dk