Human Metabolome Database (HMDB)

The HMDB Suite of Integrated Databases

The HMDB is the flagship database within a larger, interconnected suite of resources, each with a specialized focus:

• DrugBank: This provides equivalent information on approximately 2,832 drugs and 800 drug metabolites. DrugBank is a comprehensive, freely accessible online database that uniquely combines detailed drug data with extensive drug target information. Often described as a "pharmacopeia of the future," it serves as a critical bioinformatics resource for pharmacology, drug discovery, and precision medicine by linking chemical, pharmaceutical, genetic, and molecular data all in one place.
• T3DB (The Toxin and Toxin Target Database): This contains data on about 3,670 common toxins and environmental pollutants. T3DB (The Toxin and Toxin Target Database): is a freely accessible, comprehensive bioinformatics resource that is specifically dedicated to toxins and their biological targets. It serves as a detailed encyclopedia for toxic substances—from environmental chemicals and pesticides to venoms and microbial toxins—and systematically documents how they interact with the human body at the molecular level.
• SMPDB (The Small Molecule Pathway Database): It houses pathway diagrams for approximately 132,335 human metabolic, drug, and disease pathways, plus thousands more for other organisms. SMPDB: The Small Molecule Pathway Database, or SMPDB, is an essential, visually-oriented component of the HMDB suite. Its primary purpose is to map and illustrate the complex network of biochemical pathways within the human body, with a unique focus on the small molecules at the center of these processes.
• FooDB: It serves as a comprehensive resource on roughly 70,000 food components and food additives. FooDB is the world's largest and most comprehensive freely available database dedicated to the chemistry and biology of food. It serves as a detailed encyclopedia for both nutrients and flavor compounds, cataloging the thousands of biochemicals found in natural and processed food products.

Navigation tools in HMBD

The HMDB offers a variety of user-friendly tools to access its data:

• Browse Function: Provides a tabular overview of the database, allowing users to casually scroll, sort, and click on any entry for full details. A key feature under this menu is the Biospecimen link, which lists normal and abnormal metabolite concentrations for 23 different biological sample types (e.g., blood, urine, cerebrospinal fluid).
• Text and Chemical Search: Users can perform simple or advanced text queries. The ChemQuery tool allows searching by drawing a chemical structure or entering a SMILES string.
• Sequence Search: Supports BLAST searches against the database's collection of over 8,000 gene and protein sequences.
• Advanced Search: An easy-to-use yet powerful relational query tool that allows for complex searches across multiple data fields simultaneously.

Specialized Spectral Search Tools (for Experimental Data)

A defining strength of the HMDB is its direct support for experimental metabolomics data analysis:

• MS Search: Users can submit mass spectral files (in MoverZ format) to search against HMDB's library of LC-MS/MS spectra. This is used to identify metabolites from complex mixtures.
• NMR Search: Users can submit peak lists from 1D or 2D NMR spectra (such as 1H, 13C, TOCSY, or HSQC) to compare against HMDB's NMR spectral libraries for metabolite identification.

Other Resources and Information in HMBD

• Downloads section: This section provides bulk access to sequence, image, and text files from the database.
• Human Metabolome Library (HML): A linked service where researchers can order physical samples of metabolite standards for use in their own experiments.
• About Section: The HML lists metabolites that can be ordered for a fee by researchers around the world. There is About menu under which there are links for valuable metadata, including detailed database statistics and descriptions of the sources used to assemble the HMDB.

Methodology to use the Human Metabolome Database (HMDB)

Here is a step-by-step workflow for effectively using HMDB, illustrated with a practical example. This structured workflow transforms HMDB from a simple lookup table into a powerful discovery platform for metabolomics research

Case Scenario/research question: In a blood plasma sample from a patient, an unknown compound with a mass of 180.0634 Da via mass spectrometry is detected. Identify the sugar related metabolite and its biological role.

Steps to navigate Human Metabolome Database (HMDB)

Step 1: Define Your Goal and Choose the Appropriate Search Tool

First, determine what you already know about your compound and select the HMDB tool designed for that type of data.

For the given case/example: We have an accurate mass from an MS experiment. The most direct and powerful approach is to use the MS Search Alternatively, we could start with an Advanced Search using the mass value.

Step 2: Execute the Search and Filter Results

Navigate to the chosen search tool from the HMDB homepage and input your data.

Workflow A: Using MS Search (Direct Spectral Matching)

1. Click "Search" → "MS Search".
2. Upload your mass spectrum file (in supported formats like .mzXML, .mzML) or enter the observed m/z value (180.0634) and adduct type (e.g., [M+H]+, [M+Na]+).
3. Set tolerance parameters (e.g., 5 ppm mass accuracy).
4. Run the search. The database compares your data against its library of experimental and theoretical spectra.

Workflow B: Using Advanced Search (Query by Properties)

1. Click "Search" → "Advanced Search".
2. In the query builder:
• Select "Monoisotopic Molecular Weight".
• Set the operator to "Between" and enter 180.0594 and 180.0674 (allowing ± 0.004 Da/~22 ppm tolerance).
• (Optional) Add a second criterion: "Biospecimen" → "Contains" → "Blood".
3. Run the query.

Example Result: Both searches will return a primary, high-confidence hit: HMDB0000122 - D-Glucose. You will also see other isomers like galactose (same mass) but ranked lower if spectral matching is used.

Step 3: Analyze the MetaboCard Entry

Click on the HMDB ID (e.g.,HMDB0000122 ) to open the full MetaboCard. This is the core information hub. Systematically review the following sections:

• Chemical Data: Confirm structure, formula (C₆H₁₂O₆), and chemical properties.
• Clinical Data: This is crucial for our example
• "Normal Concentrations": See that normal fasting blood glucose is ~4.0-5.5 mM.
• "Abnormal Concentrations":Find direct links to hyperglycemia and diabetes mellitus. This confirms the clinical relevance of your finding.
• Biological Data:
• Pathways: Click the "Pathway Browser" link or see the "Pathway Name" field. You'll find glucose in central pathways like "Glycolysis", "Gluconeogenesis", and "Insulin signaling".
• Associated Enzymes/Genes: See links to proteins like hexokinase (HK1, HK2) and glucose transporters (SLC2A1/GLUT1, SLC2A4/GLUT4).

Step 4: Leverage Connected Databases (The HMDB Suite)

Use the hyperlinks and integrated databases to expand your investigation.

1. Navigate to SMPDB for Pathways: From the glucose MetaboCard, click the link to the "Glycolysis" pathway in SMPDB. This provides a visual, interactive diagram showing glucose's position, connected metabolites, and regulating enzymes—offering immediate biological context.
2. Check DrugBank for Therapeutics: A quick search in the integrated DrugBank (via the suite menu) for "glucose" will show it is not only a metabolite but also a drug (e.g., dextrose injection) used to treat hypoglycemia.
3. Explore FooDB for Dietary Sources: Link to FooDB to see the natural abundance of glucose in thousands of foods, completing the picture from diet to metabolism.

Step 5: Access Experimental Data and External Validation

Utilize HMDB's specialized data for experimental validation and further linking.

• Spectral Evidence: In the MetaboCard, examine the "MS/MS Spectra" and "NMR Spectra" sections. You can download these to compare directly with your own lab data for definitive confirmation.
• External Databases: Use the hyperlinks in the "General References" and "Database Links" sections (e.g., PubChem, KEGG, ChEBI) to cross-validate information and access even more literature and data.

Step 6: Download and Conclude

• For Reporting: Use the "Download" options to obtain high-resolution images of the chemical structure, pathway maps, and spectral data for your publication or report.
• Final Synthesis for Our Example: You have successfully:
  1. Identified the unknown peak as D-Glucose.
  2. Confirmed its elevated level is clinically significant for diabetes.
  3. Mapped its role in key metabolic pathways.
  4. Accessed its reference MS/NMR spectra for future comparisons.
  5. Explored its connections to drugs and diet.

Table: Summary of differences between METLIN and HMDB