pydca

pydca infers coevolutionary couplings from multiple sequence alignments to identify residue–residue interactions for structural and functional analysis.

Key Features:

• Algorithmic approaches: Implements inverse statistical methods—mean-field approximation and pseudo-likelihood maximization—to infer coevolutionary relationships between residues from MSAs.
• Input formats: Accepts multiple sequence alignment (MSA) files in FASTA format.
• Reference sequence mapping: Optionally maps a supplied reference sequence onto the MSA to compute coevolutionary scores for pairs of sites in the reference.
• Preprocessing and visualization: Provides MSA trimming and contact map visualization to process alignments and display inferred residue contacts.

Scientific Applications:

• Coevolutionary coupling inference: Identifies direct residue couplings indicative of evolutionary constraints and potential physical contacts.
• Contact and structure prediction: Supports prediction of protein and RNA contacts useful for tertiary structure modeling.
• Interpretation of sequence variation: Aids analysis of functional implications of sequence variants through changes in coupling patterns.
• Support for experimental structure determination: Provides contact information that can guide experimental efforts such as mutagenesis and structure solving.

Methodology:

Applies mean-field approximation and pseudo-likelihood maximization to MSAs in FASTA format, with optional reference-sequence mapping, and includes MSA trimming and contact map visualization.