Metadata is structured information that describes, explains, locates, or otherwise makes it easier to retrieve, use, or manage an information resource. Metadata is often called data about data or information about information. It ensures that the context for how your data was created, analyzed and stored, is clear, detailed and therefore, reproducible.
National Information Standards Organization, 2004

Biomedical Metadata

Good metadata enables you to understand, use, and share your own data now and in the future and helps other researchers discover, access, use, repurpose, and cite your data in the long term. It also facilitates long-term archival preservation of the data. Types of documentation to capture biomedical data include:

  • Reagent Metadata: Information about the clinical samples, biological reagents (e.g., cell lines, antibodies, siRNAs), chemical reagents (e.g., drugs), etc. used to generate the data.
  • Technical Metadata: Information automatically generated by research instruments and associated software.
  • Experimental Metadata: Information about the experimental conditions (e.g., assay type, time points), the experimental protocol, and the equipment used to generate the data.
  • Analytical Metadata: Information about data analysis methods, including software name and version, quality control parameters, and output file type details.
  • Dataset Level Metadata: Information about the objectives of the research project, participating investigators, relevant publications, and funding sources.
     

When to Record Metadata

Many fields within the biomedical science community are developing standards for what metadata to collect across different data types. Whenever possible, it is best to consult community standards before you begin collecting research data. It is easiest and most efficient to record metadata during the research process while the data still are active. This also ensures that the metadata record is complete and accurate.

Your metadata will likely come from several sources during your research:

  • Technical Metadata: Generated from research instruments and software used.
  • Additional Metadata: Most metadata will be collected manually. Consider using an existing schema or templates to make the process standardized and easier.
     

How to Record Metadata

Experimental and data analysis metadata should be stored alongside your research data. This documentation can be maintained in a variety of forms, including, but not limited to:

  • Lab Notebook: For research groups that use them, lab notebooks are often the primary record of the research process. They are used to document hypotheses, experiments, analyses, and interpretations of experiments. Electronic Lab Notebooks offer several advantages over traditional paper notebooks. Find an ELN solution for LMA groups.
  • Protocol: A protocol describes the procedure(s) or method(s) used in the implementation of a research project or experiment. If you need to maintain protocols, we strongly recommend a tool like protocols.io. Learn more about protocols.io for HMS/HSDM/HSPH researchers.
  • README: A README File is a text file located in a project-related folder that describes the contents and structure of the folder and/or a dataset so that a researcher can locate the information they need.
  • Data Dictionary: Also known as a codebook, a data dictionary defines and describes the elements of a dataset. This information may be similar to what you may be requested to provide when submitting data to a repository. 
     

Metadata Standards & Guidance

Where possible, employ one or multiple established metadata standards, or schemas, that are widely used within your discipline. If you are storing your data in a repository, you also must comply with its metadata requirements.

Quick help links

Subject-Specific Guidance

Clinical Data

Researchers are urged to use standardized terminology wherever possible

Clinical metadata may include elements that pose a risk of patient identification. It is the institutions' responsibility to provide the appropriate privacy training for handling this information. Explore the NIAID GSCID/BRC Clinical Metadata Standard.

Consider consulting and using the following codes, identifiers, and terminologies in your data, as applicable:

Common Data Elements and Standards

  • CDEs provide a way to standardize data collection so that related data can be pooled and analyzed across multiple studies or to investigate relationships between data in unrelated datasets
  • NIH Common Data Elements (CDE) Repository
     

Patient Identifiers

  • Social Security Number
  • Taxpayer Identification Number
  • National Provider Number
     

Diagnosis and procedure codes

Drug and device codes

Experimental Biomedical Research

Metadata about scientific experiments are essential for finding, retrieving, and reusing data

There are several current projects at LMA (many NIH-funded) that are focused on developing recommended metadata standards for describing experimental biomedical research data. In addition to the mentioned guidelines, biomedical researchers are also encouraged to consult FAIRSharing.org, an educational resource, and portal to metadata standards, databases, and data policies for a variety of disciplines.

Please Note: When describing experimental biomedical research data, it is important to identify not only the canonical reagents but also the actual batches of those reagents that were used to create your data.

  • A canonical reagent is the ideal of the reagent, and its definition and description are true for all examples of that reagent
  • Batches are the physical lots (daughters) of the canonical reagent, and there is often slight variation between batches
     

NIH LINCS

“The LINCS project is based on the premise that disrupting any one of the many steps of a given biological process will cause related changes in the molecular and cellular characteristics, behavior, and/or function of the cell – the observable composite of which is known as the cellular phenotype. Observing how and when a cell’s phenotype is altered by specific stressors can provide clues about the underlying mechanisms involved in perturbation and, ultimately, disease.”

  • LINCS metadata standards
    • These metadata standards were developed to describe LINCS reagents, assays, and experiments. They provide guidance for required, required if applicable, and optional elements.
       

Illuminating the Druggable Genome (IDG) Consortium

“The goal of the Illuminating the Druggable Genome (IDG) program is to identify and provide information on proteins that are currently not well studied within commonly drug-targeted protein families.”

Proteomics Data

Metadata is crucial to interpret and reanalyze deposited datasets

The following resources provide metadata recommendations for proteomics, interactomics, and metabolomics research:

Additional Medical Metadata Standards

Metadata Schemas

Labeling, tagging or coding system used for recording cataloging information or structuring descriptive records

Controlled Vocabularies

Lists of predefined terms by a community or research group

Medical research and biomedical professional communities may employ controlled vocabulary standards such as:

Ontologies

Variety of controlled vocabulary that defines components and describes relationships among components

Most ontologies are used for interoperability among databases, some using (Web Ontology Language (OWL) or Resource Description Frameworks (RDF)). Here are some examples of ontologies used in biomedical research:

Technical Standards

Established norm or requirement for a repeatable technical task

Technical Standards establish norms or requirements for a repeatable technical task establishing uniform criteria, methods, processes, and practices. ISO standards are internationally agreed by experts, and create a formula that describes the best way of doing something.