RRID (Research Resource Identifier)

An RRID (Research Resource Identifier) is a unique, citable identifier assigned to a specific research resource — an antibody, cell line, model organism, or software tool/database — so that the resource can be unambiguously identified and traced from a published methods section back to a single, machine-resolvable record. RRIDs were developed through the Resource Identification Initiative, a working group convened under FORCE11 (the Future of Research Communications and e-Scholarship) starting with a 2011 workshop under the Linking Animal Models to Human Diseases (LAMHDI) consortium and formalized at planning meetings in 2012 and at NIH in June 2013. The identifiers themselves, and the portal used to look them up or mint new ones, are maintained by SciCrunch’s Resource Identification Portal, hosted at the University of California, San Diego.

Operational definition

A string qualifies as a real RRID if it takes the form RRID: followed by an accession number issued by one of the authoritative source databases the initiative recognizes, and that accession number resolves — via the SciCrunch resolver or a compatible resolver such as n2t.net — to a single, curated record describing exactly one resource. The prefix segment of the accession number signals the resource type and its source database:

• Antibodies — RRID:AB_ followed by a number, sourced from the Antibody Registry (e.g. RRID:AB_90755).
• Cell lines — RRID:CVCL_ followed by a code, sourced from Cellosaurus (e.g. RRID:CVCL_1H60).
• Model organisms — species-specific accession numbers from the relevant organism database, such as MGI for mouse, ZFIN for zebrafish, RGD for rat, or the International Mouse Strain Resource (IMSR) for mouse strains.
• Software, databases, and other tools — RRID:SCR_ followed by a number, sourced from the SciCrunch registry itself (e.g. RRID:SCR_004098 for SciCrunch’s own listing).

A bare catalog number, a vendor product page URL, or a resource name typed in free text does not qualify as an RRID, even if it uniquely identifies the resource to a human reader — the defining property of an RRID is that it is machine-resolvable through a common resolution layer, free to generate and to look up, and stable across publishers rather than tied to any one journal’s own house style.

Why RRIDs exist

The Resource Identification Initiative was built to close a specific, well-documented reproducibility gap: methods sections routinely describe key research materials ambiguously — “an anti-TH antibody” or “HeLa cells” without a lot number, supplier, or authentication record — making it difficult or impossible for another lab to determine whether they used the same reagent, let alone to detect when a widely-used resource (a misidentified or cross-contaminated cell line is the canonical example) has quietly propagated errors across the literature. An RRID solves this by giving the resource itself a persistent, unambiguous, citable identity independent of how any single paper happens to describe it in prose, in the same way a DOI gives a publication one and an ORCID iD gives a researcher one. Because RRIDs are consistent across publishers, a reader, curator, or automated tool can trace every paper that cites a given antibody, cell line, or software tool regardless of which journal published it.

How to look up or register an RRID

SciCrunch’s Resource Identification Portal is the central place to search for an existing RRID before assuming one doesn’t exist, and to register a resource that doesn’t yet have one. A researcher preparing a manuscript typically: (1) searches the portal by resource name, vendor, or catalog number; (2) confirms the returned record matches the specific resource actually used (same clone, same catalog number, same species where relevant); and (3) copies the resolvable RRID: string into the manuscript at first mention of that resource, alongside the vendor and catalog information the resource type calls for. If no record exists — common for newly deposited software or a newly generated antibody — the portal supports self-registration, after which the new RRID resolves the same way as an established one.

Citation format by resource type

The Resource Identification Initiative’s guidance calls for slightly different citation detail depending on resource type, always with the RRID included at first mention in the methods section:

• Antibodies: vendor, catalog number, and RRID — e.g. “a polyclonal antibody against tyrosine hydroxylase (Chemicon, Cat. AB1542, RRID:AB_90755).”
• Cell lines: standard short name, provider, catalog or clone number, and the Cellosaurus-based RRID — e.g. “Coriell Cat# GM03745, RRID:CVCL_1H60.”
• Software tools and databases: resource name, its website where one exists, and the RRID.
• Model organisms: the RRID alone is generally considered sufficient, since the source database record already carries the strain/species detail.

Journal and funder adoption

The pilot launched in February 2014 with roughly 25 journals, including the Journal of Neuroscience, Brain and Behavior, the Journal of Comparative Neurology, and the Frontiers journals. Adoption has grown substantially since: as of 2022, over 1,000 journals had used RRIDs in published articles, with more than half a million individual RRID citations across the literature, and roughly 670 journals — including titles from Springer Nature, Cell Press, PLOS, and the American Association for Cancer Research — explicitly encourage RRID citation in author guidelines. Nature Protocols requires authors to provide RRIDs for every antibody and cell line reported. On the funder side, RRIDs sit alongside — and are one practical way to satisfy — the NIH’s broader Rigor and Reproducibility policy (in force for applications from the January 2016 due date onward), which requires an “Authentication of Key Biological and/or Chemical Resources” plan for cell lines, antibodies, and other biologics central to a proposed study; the NIH policy itself does not mandate the RRID system specifically, but citing a resource’s RRID is a straightforward, machine-verifiable way to document the authentication that policy expects.

Related terms

RRIDs sit within CASRAI’s broader persistent-identifier vocabulary alongside the DOI, ORCID iD, and ROR ID — each solves the same underlying problem (unambiguous, resolvable identity) for a different entity type: publications, researchers, organizations, and, with RRID, research materials and tools. Within reproducibility reporting specifically, RRIDs complement study-design reporting frameworks such as CONSORT 2010 (clinical trials) and ARRIVE 2.0 (animal research) — those frameworks govern how a study’s design and conduct are reported, while RRIDs govern how the specific materials used within that study are identified. See also persistent data identifier, identifier scheme, and reproducibility.