DomainMapper

DomainMapper annotates protein domain structures by identifying contiguous, non-contiguous, and insertional domains to improve the accuracy and precision of domain annotations in protein sequences.

Key Features:

• Accurate non-contiguous annotation: Assigns domain structure annotations for proteins with non-contiguous residue ranges and insertional domains.
• Comprehensive coverage: Provides annotations for both contiguous and non-contiguous domain topologies within query sequences.
• Addresses HMM limitations: Targets challenges present in traditional automated pipelines that search query sequences against hidden Markov model (HMM) profiles and often fail on non-contiguous domains.
• Computational domain identification: Uses advanced computational techniques to identify and annotate domains within query sequences.
• AlphaFold validation: Validated against the AlphaFold database, reporting that approximately 10.74% of domains in yeast and 4.52% in human proteins are non-contiguous.
• Cross-species mapping: Enables mapping of domain structures across organisms to capture topology variation.

Scientific Applications:

• Structural informatics: Provides refined domain annotations for structural analyses and interpretation of protein architectures.
• Evolutionary biology: Facilitates investigation of fold propensities for non-contiguity and insertional configurations across the Tree of Life.
• Protein function and interaction studies: Supports analysis of protein function and interactions by resolving complex domain topologies.
• Comparative domain mapping: Enables comparison of domain topology distributions across species and proteomes.

Methodology:

Applies computational techniques to identify and annotate domains within query protein sequences, addresses limitations of HMM-profile-based annotation pipelines, and was validated by comparison to the AlphaFold database.