Metabolic Atlas

Metabolic Atlas integrates genome-scale metabolic models (GEMs) across multiple species and provides curated, FAIR-aligned resources to support metabolic modeling, metabolomics, clinical chemistry, and biomarker discovery.

Key Features:

• Comprehensive Human Metabolic Models: The consensus human model Human1 consolidates curated models and supports generation of cell- and tissue-specific models using transcriptomic, proteomic, and kinetic data.
• Model Organism Coverage: Includes GEMs for Mus musculus (Mouse1), Rattus norvegicus (Rat1), Danio rerio (Zebrafish1), Drosophila melanogaster (Fruitfly1), and Caenorhabditis elegans (Worm1), incorporating orthology-based pathways and species-specific reactions.
• Enzyme Parameter Predictions (GotEnzymes): Hosts AI-predicted enzyme parameters for over 25.7 million enzyme-compound pairs across 8099 organisms.
• Human Tissue-Specific Models: Provides collections of tissue-, cell-, and cancer-specific GEMs and gut bacteria models distributed in SBML format.
• Visualization and Reaction Data: Supports visualization of metabolic networks overlaid on KEGG pathway maps and supplies biochemical reaction entries via the Human REaction Entities Database (Hreed) searchable by keyword or gene/protein references.

Scientific Applications:

• Disease Mechanism Studies: Enables integrative omics analyses to investigate metabolic alterations in diseases such as Alzheimer's disease using models like Mouse1.
• Biomarker Discovery: Facilitates comparative metabolic analyses for identification of potential biomarkers for early diagnosis.
• Systems Medicine Development: Provides interoperable GEMs to support systems-level modeling in systems medicine.
• Educational Resources: Supplies curated models and pathway visualizations for teaching metabolism and metabolic modeling.

Methodology:

Curation of existing models to produce the consensus Human1 model. Generation of cell- and tissue-specific models using transcriptomic, proteomic, and kinetic data. Use of AI to predict enzyme parameters in GotEnzymes for enzyme-compound pairs. Distribution of models in SBML format. Visualization overlaid on KEGG pathway maps. Biochemical reaction data accessible via Hreed using keyword or gene/protein searches.