HydDB

HydDB classifies hydrogenase catalytic subunit sequences to predict hydrogenase class and associated biochemical function in studies of H2 metabolism.

Key Features:

• Expanded Classification Scheme: An expanded scheme enumerating 29 [NiFe], 8 [FeFe], and 1 [Fe] hydrogenase classes, including 11 newly defined classes with distinct biological functions.
• Sequence Prediction Capability: Combines a k-nearest neighbors algorithm with Conserved Domain Database (CDD) referencing to classify hydrogenase primary amino-acid sequences and predict function, demonstrated with 12 bacteria, archaea, and eukaryotes.
• Extensive Database: Curated database of 3,248 annotated hydrogenase catalytic subunit amino-acid sequences with physiological, biochemical, and structural annotations for 38 defined hydrogenase classes.

Scientific Applications:

• Microbial Metabolism: Classification and functional prediction of hydrogenases to investigate microbial H2 metabolism.
• Evolutionary Biology: Comparative analyses of hydrogenase diversity and evolutionary history across microbial taxa.
• Biogeochemical Cycles: Inference of enzyme roles in ecosystem-level H2 cycling and biogeochemical processes.
• Genomic Studies: Prediction of hydrogenase content and function from primary sequence data in newly sequenced genomes and proteomes.

Methodology:

Classification uses a k-nearest neighbors algorithm combined with Conserved Domain Database (CDD) referencing against a curated set of 3,248 annotated hydrogenase catalytic subunit amino-acid sequences and an expanded classification scheme (29 [NiFe], 8 [FeFe], 1 [Fe]).