SweepCluster: A SNP clustering tool for detecting gene-specific sweeps in prokaryotes

Area of Science:

• Evolutionary biology
• Genomics
• Microbial genetics

Background:

• Gene-specific sweeps are crucial for microbial adaptation and divergence.
• Existing methods for detecting selective sweeps are not optimized for prokaryotic data.
• Challenges include low resolution of sweep regions and ignoring mutation distribution.

Purpose of the Study:

• To develop a novel, gene-centric, and spatial-aware method for identifying gene-specific sweeps in prokaryotes.
• To address limitations of current methods in prokaryotic genotype data analysis.
• To provide a computational tool for evolutionary insights and genetic marker discovery.

Main Methods:

• Proposed a gene-centric and spatial-aware approach for detecting gene-specific sweeps.
• Implemented the method in a Python tool named SweepCluster.
• Utilized pre-selected polymorphisms based on genetic signatures (population subdivision, linkage disequilibrium, phenotype association).

Main Results:

• SweepCluster demonstrated high sensitivity and specificity (>90%) in simulations.
• The method effectively reduced uninformative signals in bacterial datasets (Streptococcus pyogenes, Streptococcus suis).
• Achieved 78% concordance with known bacterial gene-specific sweeps and recovered 80% of known eukaryotic sweep regions.

Conclusions:

• SweepCluster is a versatile tool applicable to both prokaryotic and eukaryotic genotypic data.
• It offers valuable insights into adaptive evolution by detecting gene-specific sweeps.
• The method aids in identifying genetic markers associated with biological phenotypes.