dbCAN

dbCAN annotates carbohydrate-active enzymes (CAZymes) by detecting and assigning CAZyme signature domains in protein sequences to characterize carbohydrate metabolism-related functions.

Key Features:

• Signature domain-based annotation: Identifies and assigns CAZyme signature domains to proteins for functional annotation.
• Hidden Markov models (HMMs): Represents signature domains using hidden Markov models to capture conserved sequence features.
• Family-specific HMMs: Maintains HMMs tailored to individual CAZyme families to reflect family-specific sequence characteristics.
• Curated domain definitions: Uses a database of signature domain definitions derived from searches within the Conserved Domain Database (CDD) and literature curation.
• Automated protein-dataset annotation: Performs automated annotation of protein datasets to identify CAZyme-related domains.

Scientific Applications:

• Genome annotation: Annotates CAZymes in newly sequenced genomes and proteomes.
• Metabolic capability inference: Infers organismal carbohydrate metabolism and enzymatic capabilities from annotated CAZymes.
• Biofuel and biorefining research: Identifies enzymes relevant to biomass degradation and carbohydrate processing for biofuel and biorefining studies.
• Comparative analyses: Enables comparison of carbohydrate-active enzyme content across diverse biological systems.

Methodology:

Constructs family-specific HMMs for CAZyme signature domains; searches protein sequences with these HMMs to detect and assign signature domains; defines signature domains via searches of the Conserved Domain Database (CDD) and literature curation.