The simplicity of protein sequence-function relationships.

Area of Science:

• Molecular Biology
• Biophysics
• Genetics

Background:

• The genetic architecture of proteins, determining how amino acid sequence dictates function, is often considered complex due to pervasive high-order epistatic interactions.
• Previous methods may overestimate epistasis by using reference sequences or failing to account for global nonlinearity in sequence-function relationships.

Approach:

• Developed a novel reference-free method to simultaneously estimate global nonlinearity and specific epistatic interactions across a protein's genotype-phenotype map.
• This approach offers a more robust and efficient explanation of protein genetic architecture, resilient to noise and model limitations.

Key Points:

• Reanalysis of 20 mutagenesis experiments reveals additive and pairwise effects, plus a simple nonlinearity, explain a median of 96% of phenotypic variance.
• Third- and higher-order epistasis strongly affects only a small fraction of genotypes.
• Protein genetic architecture is sparse, requiring significantly fewer terms than genotypes to explain variance.

Conclusions:

• The sequence-function relationship in most proteins is far simpler than previously assumed.
• This finding enables more tractable and effective strategies for characterizing protein function from sequence.