SSIPe

SSIPe predicts binding affinity changes for protein-protein interactions by integrating evolutionary profiles derived from structural and sequence homology searches with a physics-based energy function.

Key Features:

• Integration of Evolutionary Profiles and Physical Energy Functions: Combines evolutionary profiles from structural and sequence homology searches with a physics-based energy function to estimate binding affinity changes for protein-protein interactions.
• Amino Acid-Specific Pseudocounts: Implements amino acid-specific pseudocounts to refine interface profiles by incorporating prior mutation probabilities for each amino acid.
• Performance Evaluation: Validated on 2,204 mutations across 177 proteins with less than 30% sequence identity between training and test datasets, achieving a Pearson correlation coefficient of 0.61 and a root-mean-square-error of 1.93 kcal/mol.
• Advantages Over Traditional Methods: Outperforms previous profile-based methods such as BindProfX by combining physics-based energy calculations with enhanced sequence-derived interface profiles and optimized pseudocounts.

Scientific Applications:

• Protein Function Annotation: Quantifying mutation-induced binding affinity changes to support annotation of protein functions based on interaction capabilities.
• Genetic Disease Diagnosis: Assessing impacts of specific mutations on protein-protein interactions to aid interpretation of genetic variants linked to disease.

Methodology:

Gathering interface profiles through structural and sequence homology searches; applying amino acid-specific pseudocounts to refine these profiles; and computing binding affinity changes using a physics-based energy function.