MDR database

MDR database catalogs sequence, structure, and functional annotations for medium-chain dehydrogenases/reductases (MDRs) to support analysis of sequence-structure-function relationships.

Key Features:

• Comprehensive dataset: Contains 2,684 MDR sequences and 42 experimentally determined structures.
• Classification: Organizes MDRs into 199 homologous families and 29 superfamilies, distinguishing 25 zinc-containing and 4 non-zinc-containing superfamilies.
• Annotated multiple sequence alignments: Provides multiple sequence alignments with annotations of functionally relevant residues for each family.
• QSDL analysis: Performs detailed analysis of the quaternary structure determining loop (QSDL), defining short, medium, and long QSDL subclasses predictive of oligomeric state (short → tetramers, long → dimers, medium → either).
• Substrate-binding site characterization: Identifies conserved substrate-binding regions across zinc-containing MDR superfamilies and two variable substrate recognition sites (SRS1 on QSDL and SRS2 in the catalytic domain for long QSDL classes).
• Sequence-structure-function integration: Integrates sequence, structural, and functional information to map residues relevant for substrate recognition and potential engineering targets.

Scientific Applications:

• Sequence-structure-function analysis: Enables systematic investigation of relationships between sequence variation, conserved tertiary structure, and enzyme function in MDRs.
• Evolutionary and functional classification: Supports analysis of evolutionary relationships and functional diversification across 199 homologous families and 29 superfamilies.
• Oligomeric state prediction: Uses QSDL length subclasses to predict MDR oligomeric states (tetramer versus dimer).
• Engineering target identification: Guides identification of functionally relevant residues and hotspots for modifying substrate specificity or other enzymatic properties.

Methodology:

Annotated multiple sequence alignments, systematic analyses to classify sequences into 199 homologous families and 29 superfamilies, annotation of functionally relevant residues, and analysis of the QSDL loop length in relation to oligomeric state.