GASS

GASS identifies active site 3D templates in protein structures to improve protein function prediction by detecting conserved catalytic and substrate-binding residues.

Key Features:

• Active site 3D template search: Locates three-dimensional arrangements of residues corresponding to active sites in protein structures.
• Heuristic conservation-based search: Uses a heuristic approach that focuses on the higher conservation of active site residues compared to other residues.
• Targeted regions: Specifically targets catalytic sites and substrate binding sites as active site components.
• Non-exact amino acid matching: Allows non-exact matches that accommodate conservative amino acid substitutions.
• Unconstrained active site size: Does not impose restrictions on the size of the active site.
• Inter-chain residue mapping: Can locate relevant amino acids that occur across different chains within a protein structure.
• Performance on CSA datasets: Demonstrated over 90% correct identification of active site templates on datasets catalogued in the Catalytic Site Atlas (CSA).

Scientific Applications:

• Protein function prediction for Pfam entries: Applied to predict functions for proteins, addressing uncharacterized entries in Pfam.
• Validation against Catalytic Site Atlas (CSA): Tested and validated using CSA-curated active site datasets.
• Comparative method evaluation: Compared with amino acid pattern search for substructures, motif identification, and catalytic site identification methods.
• CASP 10 substrate binding site evaluation: Evaluated in the CASP 10 substrate binding sites prediction competition and ranked fourth among 18 methods.

Methodology:

Applies a heuristic, conservation-focused search for catalytic and substrate-binding residues, permits non-exact (conservative) amino acid matches, places no size constraint on active sites, and can map residues across different chains.