PatchD

PatchD identifies clusters of coevolving residues in protein multiple sequence alignments to detect covarying residue patches and reveal structural and functional interaction networks.

Key Features:

• Covariation Detection: Calculates information statistics on protein sequence alignments to identify coevolving residues and residue clusters (patches).
• Enhanced Predictive Power: Extends correlated mutation analysis (CMA) beyond pairwise site comparisons by recognizing that covariation at two sites often extends to neighboring sites, forming covarying patches.
• Structural Insights: Identifies patches that frequently correspond to residues proximal in three-dimensional structure, enabling inference of potential functional interaction partners.
• Large-Scale Analysis Capability: Performs searches for coevolving regions across protein domains and extensive datasets.

Scientific Applications:

• Structural Biology: Maps covarying residue patches onto protein structures to infer regions involved in structural stability and intermolecular contacts.
• Functional Genomics: Identifies residue clusters linked to conserved functional constraints across homologous proteins in MSAs.
• Protein Interaction Prediction: Highlights residue patches that suggest intramolecular or intermolecular interaction networks and functional interfaces.
• Drug Discovery: Characterizes structurally and functionally relevant residue clusters that can inform targeted intervention strategies.

Methodology:

Analyzes multiple sequence alignments (MSAs) of proteins using information-theoretic statistics to measure covariation between residues and between disconnected sequence clusters, detecting covarying patches beyond pairwise correlated mutation analysis (CMA) and identifying individual interacting pairs and broader interaction networks.