A knowledge-based scoring function to assess quaternary associations of proteins

Area of Science:

• Computational biology
• Structural bioinformatics
• Molecular interactions

Background:

• Understanding protein-protein interactions is crucial for deciphering cellular processes.
• Experimental methods for identifying these interactions are costly and time-consuming.
• Computational approaches are essential for large-scale prediction of protein complex formation.

Purpose of the Study:

• To develop a novel computational scheme for scoring protein-protein interactions.
• To accurately predict native protein quaternary assemblies.
• To provide a reliable tool for assessing interaction viability in silico.

Main Methods:

• Developed PIZSA (Protein Interaction Z-Score Assessment), a binary classification scheme.
• Utilized statistical potentials incorporating residue-residue contact preferences and atomic contributions.
• Accounted for steric clashes in scoring protein complex conformations.

Main Results:

• PIZSA accurately discriminates between native and non-native protein conformations.
• The method outperforms 31 other scoring schemes, ranking among the top 6.
• PIZSA successfully distinguishes biological interactions from crystallization artifacts.

Conclusions:

• PIZSA offers a robust and accurate method for predicting protein-protein interactions.
• The developed scoring scheme enhances the understanding of molecular-level cellular processes.
• PIZSA is available as a web server and standalone package for broader accessibility.