Estimating linkage disequilibrium and selection from allele frequency trajectories

Area of Science:

• Evolutionary biology
• Genetics
• Bioinformatics

Background:

• Genetic sequence data collected over time offers insights into natural selection.
• Accurate measurement of selection requires accounting for linkage disequilibrium.
• Short-read sequencing limits direct measurement of linkage disequilibrium.

Purpose of the Study:

• To develop a method for estimating linkage disequilibrium from time-series allele frequencies.
• To enable linkage-aware inference for studying natural selection using allele frequency data.
• To improve the accuracy of inferring fitness effects of mutations.

Main Methods:

• Developed a method to reconstruct linkage disequilibrium from time-series allele frequencies.
• Combined reconstructed linkage information with inference methods for fitness effects.
• Introduced regularization methods based on random matrix theory for limited sampling.

Main Results:

• The developed method reliably outperforms inference that ignores linkage disequilibrium.
• Performance is comparable to using true linkage information with sufficient sampling.
• Regularization methods enhance performance under limited sampling conditions.

Conclusions:

• The method allows for linkage-aware inference using only allele frequency time series.
• This advances the study of natural selection in scenarios with limited direct linkage data.
• Facilitates more accurate estimation of mutation fitness effects.