EVcouplings

EVcouplings infers residue-level protein interactions and predicts interaction structures across proteomes using sequence coevolution.

Key Features:

• Sequence Coevolution Analysis: Utilizes evolutionary couplings derived from protein and RNA sequence alignments to predict residue-level interactions.
• Alignment Generation: Performs alignment generation from sequence data for subsequent coupling analysis.
• Model Parameter Inference: Infers model parameters required to derive evolutionary couplings from alignments.
• Proteome-Scale Application: Applies across entire proteomes, with 53% of protein pairs in Escherichia coli reported as eligible for analysis given current genomic data.
• Structural Prediction: Predicts structures of protein–protein interactions, exemplified by identification of 620 likely interactions in the Escherichia coli cell envelope and an expansion of the known interaction space by 529 pairs.
• Functional Annotation and Mutation Effect Prediction: Predicts how mutations may affect protein structure and function for functional genomics and evolutionary studies.

Scientific Applications:

• Protein Interaction Discovery: Facilitates large-scale discovery of protein–protein interactions from sequence data.
• Structure Prediction: Provides atomic- or residue-level interaction structure predictions to inform mechanistic hypotheses.
• Functional Annotation and Mutation Analysis: Supports annotation of protein function and assessment of mutation effects on structure and interaction dynamics.

Methodology:

Derives evolutionary couplings from protein and RNA sequence alignments, performs alignment generation, infers model parameters, and predicts interaction structures at residue-level resolution.