PRISM

PRISM predicts protein-protein interactions (PPIs) and generates structural models of protein complexes by matching two input protein structures to template interfaces using a knowledge-based algorithm that integrates structural similarity and evolutionary conservation.

Key Features:

• Knowledge-Based Prediction Algorithm: Uses existing structural knowledge to predict PPIs by comparing input structures to known interfaces.
• Integration of Structural Similarity and Evolutionary Conservation: Combines structural similarity metrics with evolutionary conservation at template interfaces to improve prediction fidelity.
• Template Interface Matching: Identifies matching template interfaces from a template repository to enable construction of complex models when templates are available.
• Repository of Predicted Models: Stores predicted complex structures and associated template-interface mappings for reuse and reference.
• Interface Residue Extraction: Identifies and reports interface residues from modeled complexes for downstream analysis.

Scientific Applications:

• Structural Biology: Produces atomic-scale models of protein complexes to inform structural characterization of PPIs.
• Molecular Biology: Supports analysis of the structural basis of protein function and molecular mechanisms involving PPIs.
• Drug Discovery: Enables design and evaluation of therapeutics targeting specific protein interfaces by providing modeled interaction structures.
• Disease Mechanism Elucidation: Aids investigation of how altered PPIs contribute to biological processes and disease phenotypes.

Methodology:

A knowledge-based algorithm integrates structural similarity with evolutionary conservation at template interfaces to identify matching templates from a repository and construct structural models of protein complexes from pairs of input protein structures.