Minus the Error: Testing for Positive Selection in the Presence of Residual Alignment Errors

Area of Science:

• Evolutionary biology
• Genomics
• Bioinformatics

Background:

• Positive selection drives the increase of advantageous mutations, crucial for understanding phenotypic diversity and novel traits.
• Genome-wide comparative genomics enables systems-level insights into evolutionary dynamics.
• Automated pipelines for genome-scale data can introduce sequencing, annotation, and alignment errors, impacting positive selection inference.

Purpose of the Study:

• To develop a method for detecting positive selection on protein-coding sequences that also identifies alignment errors.
• To improve the accuracy and biological relevance of positive selection inference in genome-wide studies.

Main Methods:

• Introduced BUSTED-E, an enhanced version of the flexible branch-site random effects model (BUSTED).
• BUSTED-E incorporates an 'error-sink' component to model an abiological evolutionary regime, accounting for alignment errors.
• Applied BUSTED-E to genome-scale biological datasets pre-filtered with automated alignment tools.

Main Results:

• BUSTED-E effectively identifies pervasive residual alignment errors in genome-scale datasets.
• The method yields more realistic estimates of positive selection, reducing bias.
• Improved biological interpretation of evolutionary processes is achieved.

Conclusions:

• BUSTED-E serves as a more stringent filter for identifying positive selection in genome-wide contexts.
• The model facilitates further characterization and validation of biologically significant evolutionary events.
• This approach enhances the reliability of evolutionary analyses using large-scale genomic data.