Leveraging FracMinHash Containment for Genomic d N / d S

Area of Science:

• Genomics
• Evolutionary Biology
• Bioinformatics

Background:

• Genomic data is rapidly increasing, necessitating efficient analysis tools.
• Estimating evolutionary pressures using the dN/dS ratio is crucial for understanding adaptation.
• Traditional alignment-based dN/dS methods are computationally intensive and not scalable for large datasets.

Purpose of the Study:

• To develop a novel, alignment-free method for estimating the dN/dS ratio at the genomic level.
• To leverage FracMinHash containment for rapid and scalable evolutionary analyses.
• To compare the accuracy and scalability of the new method against traditional approaches.

Main Methods:

• Utilized FracMinHash containment, a fast similarity measure, to estimate dN/dS ratios.
• Applied the method to simulated and real-world genomic datasets, including large-scale comparisons of bacterial and archaeal genomes.
• Validated the approach by comparing results with conventional alignment-based dN/dS estimations.

Main Results:

• FracMinHash dN/dS provides accurate and scalable genomic-level estimations of evolutionary pressures.
• The method enabled pairwise dN/dS estimations for over 85,000 genomes in under 5 hours.
• Identified a novel metabolic island between specific archaeal and bacterial species, demonstrating the method's utility in discovering biological insights.

Conclusions:

• FracMinHash dN/dS is a powerful, alignment-free tool for large-scale evolutionary genomic analyses.
• The method is comparable in accuracy to traditional approaches but significantly more scalable.
• Facilitates rapid evolutionary inference and functional interpretation across diverse taxa and complex datasets.