Correlated Allele Frequency Changes Reveal Clonal Structure and Selection in Temporal Genetic Data

Area of Science:

• Evolutionary biology
• Computational biology
• Genomics

Background:

• Evolving populations with frequent beneficial mutations can maintain distinct subpopulations (clones).
• Clonal dynamics are common in microbial and viral evolution but challenging to resolve with current sequencing due to limited read lengths.
• Existing methods struggle to accurately measure linkage disequilibrium and infer clonal structure from sequencing data.

Purpose of the Study:

• To develop a novel computational method for inferring clonal structure from time-series sequence data.
• To utilize correlated allele frequency changes to overcome limitations of short sequencing reads.
• To improve the estimation of mutation fitness effects and downstream evolutionary analyses.

Main Methods:

• Developed a new method to infer clonal structure by analyzing correlated allele frequency changes over time.
• Validated the method using simulations to assess accuracy in recovering known clonal structures and linkage disequilibrium.
• Applied the method to real-world data from an *E. coli* long-term evolution experiment.

Main Results:

• Simulations demonstrated accurate recovery of underlying clonal structures and precise estimation of linkage disequilibrium.
• Application to *E. coli* data revealed novel clonal structures.
• The method significantly improved predictions of mutation effects on bacterial fitness and antibiotic resistance.
• The new method is computationally efficient, requiring substantially less time for large datasets compared to existing approaches.

Conclusions:

• The developed method provides a powerful tool for inferring clonal structures when only allele frequencies are available.
• This approach enhances downstream analyses, including fitness effect estimation and prediction of antibiotic resistance.
• The computational efficiency makes it suitable for large-scale genomic datasets in evolutionary studies.