Bait-ER: A Bayesian method to detect targets of selection in Evolve-and-Resequence experiments

Area of Science:

• Evolutionary biology
• Genomics
• Computational biology

Background:

• Experimental evolution combined with high-throughput sequencing (E&R) monitors genomic changes in populations under controlled conditions.
• Identifying adaptation signatures in E&R data requires advanced statistical methods for accurate selection detection.
• Existing methods face challenges with complex demographic models and computational demands.

Purpose of the Study:

• To present Bait-ER, a novel Bayesian approach for estimating selection coefficients from E&R experiments.
• To provide a statistically robust and computationally efficient tool for detecting adaptation.
• To address limitations in current methods for analyzing genome-wide E&R data.

Main Methods:

• Developed Bait-ER, a fully Bayesian method based on the Moran model of allele evolution.
• Incorporated overlapping generations to accommodate diverse experimental designs.
• Assessed method accuracy and precision across various demographic and experimental conditions.

Main Results:

• Bait-ER demonstrates high accuracy and precision in estimating selection coefficients, performing well in most tested scenarios.
• The method is effective even for complex adaptation trajectories beyond classical sweep models.
• Bait-ER offers significant computational advantages, outperforming existing software in speed and avoiding empirical null distribution simulations.

Conclusions:

• Bait-ER provides a powerful, accurate, and computationally efficient tool for analyzing E&R data.
• The method enhances the ability to detect selection signatures in genome-wide studies.
• An open-source package for Bait-ER is available, facilitating broader application in evolutionary research.