1  Purpose of (meta)data reporting template

Researchers in PARC are expected to contribute experimental data and associated files (e.g., images, raw data) to related / identified repositories (as outlined in PARC overarching and in project-level Data Management Plans and details provided via the DSW tool for Data Management) and to provide rich metadata1 about their datasets (to enable resource discovery and resource use/reuse) to the PARC Chemical Risk Assessment Hub (CRA Hub). The rich metadata for toxicology datasets (i.e., for data generated in PARC WP5) should describe:

  1. the test substances used in the experiments
  2. the cell lines and/or organisms (i.e., test systems) used in the experiments
  3. the protocols, guidance documents, summary of processes used for experimentation
  4. the assay methods used
  5. Summary of data processing steps
  6. details about the WP5 researchers that generated the data and their institutions.

Metadata, though its definition varies according to the context,2 is generally known as data about data. As per Gilliland (2016), metadata can be defined as:

the sum total of what one can say at any given moment about any information object at any level of aggregation. The information object can be a single item, aggregate of many items or an entire database/record keeping system.3

It is useful here to note that metadata is frequently governed by community-developed and community-fostered standards and best practices to ensure quality, consistency, and interoperability. Hence, metadata development needs collaboration, within a domain specific community to build consensus on the (meta)data scheme development and reporting template.

The overarching goal is to create a data structure, i.e., a metadata schema, which can be annotated for machine processing and to improve sharing, reuse, and retrieval of PARC generated data. The template in MS Excel contains some aspects of generalist metadata (i.e., has a few elements from the Simple Dublin Core Metadata element set) and aims to create (after reviews and broader buy-in from the PARC toxicological community) a rich, specialist metadata element set and metadata schema. Schemas help to organise and manage data in a structured, consistent, and formal way, which can help to ensure data quality and reuse. This template is an effort towards that, as a basis for refinement for the individual sub-domains and data types.

Moreover, preparing datasets with above details is a step towards FAIRification goals of PARC, which could be presented as:

The metadata requirements are intended to be embedded within the “object” (i.e., the MS Excel workbook) describing the experiment as well as stand-alone metadata (independent of where the “object” is stored and maintained). For example, the test substance and test system details, the exposure scheme, the assay, etc. will be part of the data as well as part of stand-alone metadata, this metadata could be used for relevant repositories in which the data is deposited, as well as in the PARC project’s Chemical Risk Assessment (CRA) Hub (the PARC FAIR Data Hub).

The associated template (in .xlsx) with this guide is a generic/conventional in vitro toxicity template which is to be customised as per toxicity sub-domains (e.g. genotoxicity, developmental neurotoxicity). The fields of the customised templates are being defined and tailored specifically to each toxicity domain.

The following instructions are for researchers in PARC WP5 who are generating in vitro experimental data (research outputs). The document describes the information needed in the WP5 data reporting templates for experiments and constitutes the meta(data) that is required to register all PARC WP5 datasets / research outputs in the PARC CRAHub (PARC’s metadata registry).

  1. F2 principle of FAIR: data are described with rich metadata (clear and accessible usage license, detailed provenance, and domain-related community standards).↩︎

  2. Metadata for one can be data for someone else, i.e., metadata for the data generator often becomes data for someone re-using the data. For example, metadata of a particular in vitro cytotoxicity experiment can be data for a registry which has a collection of different experiments, viz., both in vivo and in vitro experiments assessing immunotoxic effects of a particular compound.↩︎

  3. Gilliland, Anne J. (2016) “Setting the Stage.” In Introduction to Metadata, edited by Murtha Baca. 3rd ed. Los Angeles: Getty Publications. http://www.getty.edu/publications/intrometadata/setting-the-stage/. Accessed 18 December 2024↩︎