Amino acid clustering and epistasis guide sequence-based prediction of phenotype modulation

Amino acid clustering and epistasis jointly improve sequence-based prediction of quantitative phenotype modulation across protein families, with hydrophobicity emerging as crucial.

• Why it matters: Functional adaptation can reflect combined effects of multiple residues, but it remains difficult to pinpoint which sequence positions modulate quantitative phenotypes.
• What they did: The authors built an improved strategy using family-specific sequence–phenotype datasets, relating amino acid identity to experimentally measured quantitative properties and testing epistasis and amino-acid clustering.
• The result: Epistasis is crucial for phenotype modulation, and clustering improves performance by capturing physicochemical amino-acid classes; comparison with experimentally identified positions yields partial recovery plus candidate new sites.

AI summary of the abstract · may contain errors · not medical advice · read the paper.