Detecting positively selected amino acid sites using posterior predictive P-values

Area of Science:

• Evolutionary biology
• Molecular biology
• Bioinformatics

Background:

• Identifying amino acid sites under positive selection is crucial for inferring protein function.
• Sites undergoing positive selection often indicate critical roles in protein evolution.

Purpose of the Study:

• To present a novel Bayesian method for detecting positively selected amino acid sites.
• To apply this method to hemagglutinin sequences from the Influenza virus.
• To explore its utility in inferring evolutionary history.

Main Methods:

• Bayesian statistical framework for phylogenetic analysis.
• Application to hemagglutinin (HA) protein sequences from Influenza virus.
• Comparative analysis with existing methods for positive selection detection.

Main Results:

• The new Bayesian method successfully identified positively selected amino acid sites.
• Results align with those from previously established methods.
• Demonstrated that positively selected sites exhibit a tendency for conservative amino acid substitutions.

Conclusions:

• The developed Bayesian method is effective for identifying sites under positive selection.
• The method provides insights into protein function and evolutionary dynamics.
• Positively selected sites in hemagglutinin may be associated with conserved functional roles.