A phylogenetic model for the detection of epistatic interactions

Area of Science:

• Population Genetics
• Molecular Evolution
• Bioinformatics

Background:

• Paired epistatic interactions in RNA stem regions are crucial for biological functions.
• Existing statistical tools are insufficient for analyzing selection in these regions.
• Understanding selective pressures is vital for RNA structure-function relationships.

Purpose of the Study:

• To develop a novel statistical model for analyzing paired epistatic regions in RNA.
• To detect departures from selective neutrality in RNA stem regions.
• To quantify the strength of natural selection against deleterious allele combinations.

Main Methods:

• A population genetics model based on compensatory substitution processes.
• Analysis of relative rates of double and single nucleotide substitutions.
• Implementation in a fully Bayesian framework for parameter estimation.
• Application to a 5S ribosomal RNA (rRNA) dataset.

Main Results:

• The model successfully differentiates between nonindependent and negatively epistatic interactions.
• It quantifies the relative strength of natural selection acting on allele combinations.
• The approach inherently accounts for significant rate variation among RNA stem positions.

Conclusions:

• The developed model provides a robust statistical tool for studying selection in paired epistatic regions.
• This method enhances the analysis of RNA molecular evolution and functional constraints.
• The findings contribute to a deeper understanding of RNA's role in biological processes.