Explanation for each dataset field

While you are in the article Upload Dataset you may have encountered the need to understand what to write in each metadata field of your dataset. The goal of this article is to explain the relevant aspects of metadata completion in the Datasets published in CORA.RDR and the explanation and possible values to fill in each field.

As you know, each dataset in CORA.RDR contains many metadata fields that describe the dataset and other aspects related to it. It is very important to fill each field with the correct and right data. There is a template that we will use in order to provide with specific descriptions called “ICIQ_template”. The first section of the template is called “Citation Metadata” and contains the most important metadata of a dataset, which is the metadata used to cite the dataset.

Here we will add information about each field in this section and what type of information needs.

Title (required)

The main title of the dataset. It should be descriptive and consistent with related publications. If you’re uploading supplementary material for an article, use a clear title such as: Replication data for “Catalytic activation of CO₂ on Cu surfaces” Tip: Avoid abbreviations or internal references.

Usually, if you are uploading the files for a supplementary materials of an article, the title of the dataset is “Replication data for NAME_OF_THE_ARTICLE”.

Author (required)

The creator(s) of the dataset — individuals or organizations who produced or collected the data. Each author entry can include:

  • Name (required) — Personal or organizational. Format for persons: Family Name, Given Name. For example: “Mariné-Tena, Aleix”
  • Affiliation — The institution or organization to which the author belongs.
  • Identifier Type (optional) — e.g., ORCID, ISNI, VIAF, etc.
  • Identifier (optional) — the unique code (e.g. for ORCID, 0000-0002-1234-5678).

    [!CAUTION] You must provide at least one ORCID for each author. The Dataset will not be accepted for publication without this condition.

Example:

Name Affiliation Identifier Type Identifier
García, Marta Institute of Chemical Research of Catalonia (ICIQ) ORCID 0000-0003-1234-5678
Mariné-Tena, Aleix Institute of Chemical Research of Catalonia (ICIQ) ORCID 0000-0001-9273-1203

Tips:

  • Use institutional affiliation, not the project name.
  • If multiple authors, list all in their order of contribution.

Contact (required)

The person or organization users can reach with dataset questions. Each entry includes:

  • Name (required) — Person or organization.
  • Affiliation — Institutional affiliation.
  • E-mail (required) — Contact address for inquiries.

Example:

Name Affiliation E-mail
Pérez, Ana ICIQ aperez@iciq.es
Mariné-Tena, Aleix ICIQ amarine@iciq.es

If the contact is an institutional mailbox, use a functional address (e.g., data@iciq.es).

Description (required)

A summary describing the purpose, content, and scope of the dataset. Include:

  • The context of data collection,
  • The methodology used,
  • The type of data (e.g., experimental spectra, computational models),
  • And any relevant details on usage or limitations.

You can also provide:

  • Date — When the description was added or updated.

Example:

This dataset contains raw and processed data used to generate the results of the publication “Activation of CO₂ on Cu surfaces”. The dataset includes computational input/output files and experimental spectra.

Supports limited HTML tags for formatting (<p>, <b>, <i>, <ul>).

Subject (required)

Defines the broad field(s) of study related to the dataset. You can select one or more from the controlled list:

Available subjects:

  • Agricultural Sciences
  • Arts and Humanities
  • Astronomy and Astrophysics
  • Business and Management
  • Chemistry
  • Computer and Information Science
  • Earth and Environmental Sciences
  • Engineering
  • Law
  • Mathematical Sciences
  • Medicine, Health and Life Sciences
  • Physics
  • Social Sciences
  • Other

Tips:

In ICIQ datasets probably we will use Chemistry or Physics

Keyword (required)

Specific terms or phrases that describe key aspects of the dataset. Each keyword entry includes:

  • Term (required) — The actual keyword (e.g., “Catalysis”, “CO₂ reduction”).
  • Term URI (optional) — Link to a term definition or ontology (e.g., Wikidata, MeSH…).
  • Controlled Vocabulary Name (optional) — e.g., MeSH, LCSH, GEMET, CHEBI, RXNO, Wikidata…
  • Controlled Vocabulary URL (optional) — Link to the vocabulary site.

We do recommend the usage of RXNO for naming reactions and substances, ChEBI for naming proteins and biologically active proteins, MeSH for general therms and Wikidata for the rest of concepts.

Example:

Term Term URI Vocabulary Name Vocabulary URL
Catalysis https://www.wikidata.org/wiki/Q188928 Wikidata https://www.wikidata.org
Caffeine https://www.ebi.ac.uk/chebi/CHEBI:27732 ChEBI https://www.ebi.ac.uk/chebi
Archael Proteins https://www.ncbi.nlm.nih.gov/mesh/ MeSH https://www.ebi.ac.uk/chebi
Zinc atom http://purl.obolibrary.org/obo/RXNO_0000000 RXNO http://purl.obolibrary.org/obo

Tip: Choose 3–10 relevant keywords.

Used to reference any publication(s) associated with the dataset — typically journal articles that describe or use the data. Each entry includes:

  • Relation Type — Defines the relationship between dataset and publication. Options include:

    • Is Cited By
    • Cites
    • Is Supplement To
    • Is Supplemented By
    • Is Referenced By
    • References
  • Citation — Full bibliographic reference in standard citation format.
  • Identifier Type — e.g., DOI, arXiv, PMID, ISBN, URL.
  • Identifier — The specific code (e.g., 10.1038/s41586-024-00000-1).
  • URL — Link to the publication (DOI or publisher page).

Example:

Relation Type: Is Supplement To Citation: M. García et al., “Activation of CO₂ on Cu surfaces”, Nature Chemistry, 2024. Identifier Type: DOI Identifier: 10.1038/s41586-024-00000-1 URL: https://doi.org/10.1038/s41586-024-00000-1

Tips: The object and subject of the relations can be confusing. To clarify, we can add “my dataset” before the type of relation. For example: My dataset is supplement to this reference, My dataset cites this reference

Notes

Free-text field for additional information not covered elsewhere. Use it for:

  • Special acknowledgements.
  • Internal comments relevant to future reuse.
  • Description of some file structure not covered elsewhere

Example:

The data files include an additional folder with intermediate outputs not used in the final publication but relevant for reproducibility.

Language (required)

What this field means

Select the language(s) used in the dataset’s files (data tables, READMEs, code comments, documentation, questionnaires, figures’ labels, etc.). It’s not the UI language of CORA.RDR.

The widget is a multi-select list (checkbox menu). It draws from a large, controlled list of languages (ISO-style names). You may pick one or several.

What to enter

  • Choose every language that appears meaningfully in the files you are publishing:
    • README / documentation language
    • Column headers / variable labels
    • Free-text survey answers
    • Code comments / notebooks’ narrative text
    • Figure captions / axes labels
  • If the dataset mixes languages, select all that apply.
  • If the files contain no linguistic content (pure binaries, numerical arrays, images without text), use the special option “No linguistic content”.
  • If you truly cannot determine the language, use “Undetermined” as a last resort.
  • If the dataset intentionally contains many languages, you can also use “Multiple languages” (mul) where provided,

Typical selections at ICIQ

  • English
  • Catalan
  • Spanish (often combinations of the above)

Examples

Example 1 — Supplementary data with English README + Catalan column headers

  • Language: English; Catalan

Example 2 — DFT outputs (logs in English) + Jupyter notebooks with English narrative

  • Language: English

Example 3 — Spectra (binary/raw) + CSV with symbol-only columns, no words

  • Language: No linguistic content (zxx)

Example 4 — Survey data with free-text answers in Spanish and Catalan

  • Language: Spanish; Catalan

Example 5 — Large multilingual corpus with many languages

  • Language: Multiple languages (mul) (or list the top languages if guidance requires specificity)

Producer

What this field means

Affiliation is the parent or associated organization of the Producer. Use it when the Producer is a person, a lab/department, or any sub-unit that belongs to a larger body (e.g., an institute, university, consortium).

  • If the Producer is a person → Affiliation is their institution.
  • If the Producer is an organizational sub-unit (lab, center, department) → Affiliation is the umbrella organization.
  • If the Producer is already the top-level organization → Affiliation is the same top-level organization.

When it becomes required

The Producer block is conditionally required: if you enter values in one or more Producer subfields, Name becomes required for that Producer entry. Affiliation itself remains optional, but strongly recommended for clarity when Producer ≠ top-level institution.

What to enter

  • The official, canonical name of the parent organization.
  • Avoid acronyms alone if they’re ambiguous; prefer full name (you can also add the acronym in Abbreviated Name).

Formatting guidance

  • Organizations: Full Official Name (Acronym) (acronym optional)
  • Departments/Units: Parent Organization – Unit Name (or just Parent Organization in Affiliation, and Unit in Producer Name)
  • Persons: Use the organization name; do not put personal titles or roles here.

Examples

Producer is a lab (organizational sub-unit)
  • Name: Pérez Group – Heterogeneous Catalysis Lab
  • Affiliation: Institut Català d'Investigació Química (ICIQ)
Producer is a person
  • Name: García, Marta
  • Affiliation: Institut Català d'Investigació Química (ICIQ)
Producer is the top-level organization
  • Name: Institut Català d'Investigació Química (ICIQ)
  • Affiliation: Institut Català d'Investigació Química (ICIQ)
Multi-layer example (department inside a university or research center)
  • Name: Catalyst Selection and Optimization Laboratory (CSOL)
  • Affiliation: Institut Català d'Investigació Química (ICIQ)

Good practices

  • Be specific but concise: If the Producer Name already contains the lab/unit, put the umbrella institute in Affiliation.
  • Use canonical names: Match the organization’s website and publications.
  • Disambiguate with acronyms: Add acronym in parentheses to the Producer’s Abbreviated Name field, not in Affiliation (unless it’s part of the official style).
  • One entry per Producer: If you have multiple Producers, each can have its own Affiliation.

Common pitfalls (and fixes)

  • Using roles/titles like “PI”, “Professor” → Affiliation must be an organization name, not a job title.
  • Only an acronym (e.g., “ICIQ”) → Prefer “Institut Català d’Investigació Química (ICIQ)”

Distributor

Purpose: Identify who distributes/serves copies of the dataset (often the repository/collection). Diff. from Producer: Producer manages/finances creation; Distributor publishes/provides access. Multiplicity: You can add multiple distributors (primary first). Once any subfield is filled, Name is expected.

Fields to fill:

  • Name — Official distributor name (person/unit/org). Required once you add anything here. Currently, the distributor of ICIQ datasets is the data steward, so you may use Aleix Mariné-Tena for this field.
  • Affiliation — Parent org. of the distributor. Currently, the distributor of ICIQ datasets is the data steward, so you may use Institut Català d'Investigació Química (ICIQ) as the producer affiliation.
  • Abbreviated Name — Short form/acronym used in badges. In our case Aleix Mariné-Tena.
  • URL — Public landing page for the distributor. Currently, the distributor of ICIQ datasets is the data steward, so you may use https://iciq.org/staff/marine-tena-aleix/ for this field.
  • Logo URL — Direct public link to the logo image (SVG/PNG/JPG). Tip: Avoid intranet/temporary links; make sure the URL ends with the image file.

Quick tips:

  • Use official names (accents allowed), and a public, stable logo URL.
  • If Producer also distributes, list the org in both roles with appropriate URLs.
  • Keep entries concise and parallel to the Producer block to aid consistency.

Minimal examples:

  • Repository as distributor

    • Name: CORA.RDR – ICIQ Collection
    • Abbrev.: ICIQ
    • URL: https://dataverse.csuc.cat/dataverse/ICIQ
    • Logo URL: https://…/cora-rdr-iciq.svg

  • Institute as distributor

    • Name: Institut Català d'Investigació Química (ICIQ)
    • Affiliation: Centre CERCA
    • Abbrev.: ICIQ
    • URL: https://iciq.org/
    • Logo URL: (public image link, not intranet)

Deposit Date

Purpose: Record the calendar date the dataset was deposited in the repository (administrative provenance; distinct from publication or data collection dates).

Format: YYYY-MM-DD (ISO-8601). Example (yours): 2025-11-03

How to fill:

  • Use the repository’s recorded deposit date (often auto-filled on submit).
  • If entering manually, use the date when the deposit action completed (not the project end date, not the article’s publication date).

Notes & tips:

  • No time or timezone—date only.
  • This field may differ from Release/Publication Date (when the dataset becomes public) and from Data Collection Dates.
  • When you update or re-deposit a new version, the deposit date should reflect the new deposit for that version.

Common pitfalls:

  • Using acceptance or publication date → use deposit date instead.
  • Entering locale formats (e.g., 03/11/2025) → use ISO 2025-11-03.

Kind of Data

Purpose: Specify the type(s) of data included in the files. This helps users and indexers understand what they’ll download. How to fill: It’s a multi-select list. Choose all that apply to your dataset’s contents (not the research domain).

General tips

  • Be inclusive but precise: tick every category that truly appears.
  • If your dataset mixes experimental, computational, and documentation, select multiple.
  • When in doubt, prefer the most specific option available.

Option glossary (what to pick & typical ICIQ examples)

  • Survey data – Questionnaires, forms, Likert scales; CSV/TSV with respondent rows.
  • Census/enumeration data – Population/registry counts, complete listings.
  • Aggregate data – Summaries/means/totals (e.g., averaged catalytic rates by condition).
  • Clinical data – Patient/clinical measurements (ensure compliance & de-identification).
  • Event/transaction data – Time-stamped logs of actions/events (e.g., instrument events).
  • Program source code – Scripts, packages, notebooks (e.g., Python, R, MATLAB, bash).
  • Machine-readable text – Structured textual formats for automated parsing (e.g., JSON, XML).
  • Administrative records data – Institutional records (grants, inventory, instrument usage).
  • Experimental data – Raw/processed lab outputs: NMR/IR/UV-Vis/MS, XRD, electrochemistry, kinetics, HPLC/GC, etc.
  • Psychological test – Psychometrics/test batteries (rare for chemistry).
  • Textual data – Human-readable text corpora, README-heavy datasets, protocols in DOC/PDF/TXT.
  • Coded textual – Text annotated with codes/tags/categories (e.g., labeled literature corpora).
  • Coded documents – Documents with structured coding schemes or markup beyond plain text.
  • Time budget diaries – Time-use logs by participants.
  • Observation data/ratings – Human ratings/annotations of observations (e.g., microscopy scoring).
  • Process-produced – Data generated automatically by processes/systems (e.g., ELN exports, LIMS dumps).
  • Other – Use only if none of the above fits (describe in Notes).
  • Compiled data – Data collected from multiple external sources into one dataset.
  • Encoded data – Data transformed/encoded for analysis (e.g., one-hot, fingerprints).
  • Genomic data – Sequencing/omics matrices, FASTQ/FASTA, VCF, etc.
  • Geospatial data – Coordinates, shapefiles, rasters, trajectories with lat/long.
  • Laboratory notebook – ELN/PDF scans, exported notebook entries.
  • Measurement and test data – Calibration runs, instrument qualification, standard tests.
  • Recorded data – Audio/video recordings, sensor streams not otherwise classified.
  • Simulation data – Outputs from simulations (e.g., MD trajectories, DFT scans, COMSOL).
  • Images – Non-medical images: microscopy (SEM/TEM/AFM), plots (PNG/SVG), gel images.
  • Medical images – DICOM/clinical imaging (MRI/CT/etc.).
  • Video – MP4/MOV instrument captures, reaction videos, microscopy time-lapse.
  • Sound – Audio files (WAV/MP3), sonification outputs.
  • Musical composition – Scores/MIDI, not typical for science.

Quick mapping for common CORA.RDR chemistry datasets

  • Spectroscopy/Chromatography: Experimental data, Measurement and test data, (Images) for exported plots.
  • Crystallography (CIF + refinements): Experimental data, Images (diffraction/figures).
  • Electrochemistry/Kinetics: Experimental data, Measurement and test data, Aggregate data (summary tables).
  • Computational (DFT/MD): Simulation data, Program source code (scripts), Machine-readable text (JSON/XML), possibly Images (figures).
  • ELN exports: Laboratory notebook, Process-produced, Textual data.
  • Jupyter/analysis repo: Program source code, Machine-readable text, Compiled data (if multiple sources merged).

Examples (compact)

  • NMR (FIDs + processed) + GC–MS + CSV summaries: Experimental data; Measurement and test data; Images
  • DFT outputs (ORCA) + XYZ + notebooks: Simulation data; Program source code; Machine-readable text; Images
  • ELN PDF + raw instrument logs: Laboratory notebook; Process-produced; Recorded data

Common pitfalls

  • Selecting only Experimental data but forgetting Measurement and test data for calibrations/QA runs.
  • Marking Images but not Experimental data when images are derived from experiments.
  • Using Other when a suitable specific category exists—prefer the closest match and explain details in Description/Notes.