Erasing Errors Due to Alignment Ambiguity When Estimating Positive Selection

• Activity Number: 639
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Benjamin Redelings*+
• Companies: Duke University
• Keywords: Bayesian ; MCMC ; inference ; evolutionary tree ; alignment ; positive selection

Abstract

Accurate estimates of positive selection rely on also having an accurate multiple sequence alignment. Simulation studies have shown that relying on a single estimate of the alignment from commonly used alignment software can lead to unacceptably high false-positive rates in detecting positive selection. We eliminate false positives resulting from alignment error by jointly estimating the degree of positive selection and the alignment inside an evolutionary model. This model treats both substitutions and insertions/deletions as mutations on a tree, and allows site-heterogeneity in the substitution process. We conduct inference starting from unaligned sequence data by integrating over all alignments. This approach takes unaligned sequences as input, and naturally accounts for ambiguous alignments without requiring ambiguous sites to be identified and removed prior to analysis. Inference is conducted using MCMC to integrate over all alignments on a fixed evolutionary tree topology. We show using simulated data that this approach solves the problem with excessive false positives, and has nearly the same power to detect positive selection as knowing the true alignment.