A Research Resource Identifier (RRID) is a unique, citable string — RRID: followed by an accession number — that identifies a specific research resource used in a study: an antibody, a cell line, a model organism, or a piece of software or database. Where a catalog number or a resource’s plain-text name is ambiguous (vendors discontinue products, resources get renamed, the same nominal antibody clone ships under different catalog numbers from different resellers), an RRID resolves to one curated record, every time, indefinitely. This guide covers how to find or register an RRID, how to cite one correctly in a methods section for each resource type, and why an increasing number of journals ask for them. For the underlying definition and format rules, see the RRID dictionary entry; this guide focuses on the citation workflow itself.
Why methods sections need resource identifiers at all
Reproducibility failures in biomedical research are often traced not to a flawed hypothesis or a statistical error, but to a resource that could not be re-identified from the published methods section: an antibody described only as “anti-NeuN (Millipore)” with no catalog number, a cell line whose identity later turned out to be contaminated or misidentified, a software tool named without a version. A name and a vendor are not always enough — vendors merge, catalog numbers get reused for reformulated products, and free-text resource names are not machine-searchable in any consistent way. RRIDs were built to close that gap: an RRID is machine-resolvable, so a reader, a curator, or an automated screening tool can look up exactly which resource a paper used, even years later and even if the vendor or catalog number changes.
RRIDs were developed through the Resource Identification Initiative, a working group convened under FORCE11 (the Future of Research Communications and e-Scholarship). The initiative traces to a 2011 workshop under the Linking Animal Models to Human Diseases (LAMHDI) consortium, followed by planning meetings in 2012 and at NIH in June 2013, with a pilot launching in February 2014 across roughly 25 journals including the Journal of Neuroscience, Brain and Behavior, the Journal of Comparative Neurology, and several Frontiers titles. Uptake has grown substantially since: by 2022, more than 1,000 journals had published papers citing at least one RRID, with over 500,000 individual RRID citations accumulated in the literature. This growth is part of a broader push for reproducibility in published research, alongside standards such as the FAIR data principles for datasets.
The RRID format and what each prefix means
An RRID always takes the form RRID: followed by an accession number from one of a small number of authoritative source databases. The prefix segment tells you which database issued it and, by extension, what kind of resource it identifies:
- Antibodies —
RRID:AB_followed by a number, sourced from the Antibody Registry (example:RRID:AB_90755). - Cell lines —
RRID:CVCL_followed by a code, sourced from Cellosaurus (example:RRID:CVCL_1H60). - Model organisms — species-specific accession numbers issued by the relevant organism database: MGI for mouse, ZFIN for zebrafish, RGD for rat, or the International Mouse Strain Resource (IMSR) for specific mouse strains.
- Software, databases, and other tools —
RRID:SCR_followed by a number, sourced from the SciCrunch registry itself (example:RRID:SCR_004098).
A bare catalog number, a vendor product-page URL, or a resource name typed as free text does not qualify as an RRID — the identifier has to resolve through the SciCrunch resolver or a compatible resolver such as n2t.net to a single curated record. This is the same resolvability principle that underlies other persistent identifiers in research infrastructure, such as the DOI for publications and datasets or the ROR ID for institutions — a resolvable string outlives any single vendor page or URL.
How to find the RRID for a resource you’re using
Before writing a methods section, search for the resource on SciCrunch’s Resource Identification Portal (scicrunch.org/resources), which is the lookup and registration portal maintained by SciCrunch at the University of California, San Diego. The portal lets you search by resource name, vendor, or catalog number and returns the matching curated record with its RRID, if one already exists. For antibodies specifically, the Antibody Registry (a component the portal searches) is the authoritative source; for cell lines, Cellosaurus. Many vendors now list the RRID directly on their own product pages, but always confirm it against the registry rather than trusting a vendor listing at face value, since vendor-side errors do occur.
If you are reusing a resource cited in a prior paper, the RRID is often listed directly in that paper’s methods section — that’s a valid way to find it, provided the resource (e.g., the specific antibody clone and catalog number) is genuinely identical to what you’re using, not just similarly named.
Citing RRIDs in a methods section, by resource type
The Resource Identification Initiative’s own guidance is to cite the RRID immediately after the resource’s first mention in the methods section, alongside the identifying details a reader would otherwise look for. Formats vary by resource type:
- Antibodies — vendor, catalog number, and RRID at first mention: “Chemicon, Cat# AB1542, RRID:AB_90755.” Because antibody clones are frequently resold under different catalog numbers by different distributors, the RRID is what actually disambiguates the antibody itself, independent of which reseller supplied it.
- Cell lines — standard short name, provider, and catalog/clone number alongside the RRID: “Coriell Cat# GM03745, RRID:CVCL_1H60.” This matters particularly for cell lines, where misidentification and cross-contamination between lines is a well-documented, recurring problem in the literature.
- Model organisms — the RRID alone is generally sufficient, since organism-database records already carry the strain/line details a reader needs (e.g., a specific MGI or ZFIN accession).
- Software, databases, and other tools — resource name, a link to the tool’s website where one exists, and the RRID: e.g., a statistical package or an image-analysis tool cited as “[Tool name], RRID:SCR_XXXXXX.” Citing the RRID rather than only a version number lets the specific tool be traced even if the project is later renamed or the original URL goes offline.
What if the resource doesn’t have an RRID yet
Not every resource is pre-registered. If a search on the SciCrunch portal returns no match, most resource types can be registered directly through the portal — submitting the resource’s identifying details (vendor, catalog number, or for a software tool, its name and a stable URL) so SciCrunch can mint a new accession number and curate a record. Registration is free and does not require the resource’s own maker to be the one submitting it; a researcher who is the first to cite a given tool in a methods section can register it. For antibodies and cell lines specifically, registration is generally handled by submitting to the Antibody Registry or Cellosaurus respectively, both of which the SciCrunch portal indexes.
Why journals ask for RRIDs
RRID citation sits inside the broader push for rigor and reproducibility in reported methods. NIH’s policy on rigor and reproducibility (formalized in Notice NOT-OD-15-103, in effect for NIH grant applications from the January 25, 2016 due date onward) requires applicants to include a plan for authenticating key biological and/or chemical resources — cell lines, antibodies, and other biologics central to the proposed research — because such resources can drift, degrade, or be misidentified over time. It’s worth being precise here: NIH’s policy requires an authentication plan; it does not, by itself, name the RRID system as a required mechanism. RRID citation is best understood as a widely-adopted, machine-verifiable way to document exactly that kind of authentication and resource provenance, not as a literal line-item in federal grants policy.
Independent of NIH policy, a substantial share of journals now encourage or require RRID citation directly in their author guidelines — reporting as of 2022 put the figure at roughly 670 journals, spanning publishers including Springer Nature, Cell Press, PLOS, and AACR journals, with some titles (Nature Protocols among them) requiring an RRID for every antibody and cell line reported. Because journal policy changes over time and this site does not track every title’s current instructions, always check the specific journal’s author guidelines for its current RRID requirement before submission rather than assuming a blanket rule.
RRID vs. a plain catalog number
A catalog number identifies a product listing at a single point in time, at a single vendor. It can be reused, discontinued, or reassigned, and it says nothing about whether two catalog numbers from different resellers refer to the same underlying antibody clone or cell line. An RRID identifies the underlying resource itself, independent of any one vendor’s catalog, and resolves through a stable resolver rather than a vendor’s own (frequently reorganized) website. Citing both together — catalog number for immediate ordering reference, RRID for durable identification — is the pattern the Resource Identification Initiative recommends, and the one reflected in the citation examples above.
Frequently asked questions
How do I find the RRID for my antibody?
Search the vendor name, catalog number, or antibody target on SciCrunch’s Resource Identification Portal, which indexes the Antibody Registry. If the antibody is already registered, the portal returns its RRID:AB_ accession number directly.
What if my resource doesn’t have an RRID yet?
Register it through the SciCrunch portal (or, for antibodies and cell lines specifically, through the Antibody Registry or Cellosaurus). Registration is free and does not need to be done by the resource’s original maker — the first researcher to cite it in a methods section can register it.
Do I need an RRID for commonly used software?
If the software or tool has an existing RRID:SCR_ entry in the SciCrunch registry, citing it alongside the tool name is good practice and increasingly requested by journals, even for widely used software — version numbers alone don’t survive a tool being renamed or its original site going offline the way a resolvable RRID does.
How is an RRID different from a catalog number?
A catalog number is vendor- and product-listing-specific and can be reused or discontinued. An RRID identifies the resource itself and resolves through a stable resolver regardless of which vendor currently sells it or whether the original product page still exists.
Which journals require RRIDs?
Requirements vary by publisher and title and change over time, so check the specific journal’s current author guidelines. As of the most recent figures available, roughly 670 journals across major publishers (including Springer Nature, Cell Press, PLOS, and AACR titles) explicitly encourage RRID citation, and some, such as Nature Protocols, require one for every antibody and cell line reported.
Does NIH require RRIDs specifically?
Not by name. NIH’s rigor and reproducibility policy (NOT-OD-15-103) requires an authentication plan for key biological and chemical resources in grant applications; RRID citation is a practical, widely used way to satisfy that expectation, not a literally named federal requirement.
Related CASRAI resources
For the underlying definition and format rules, see the RRID dictionary entry. For how persistent identifiers work more broadly across the research-information ecosystem, see the CRIS & identifiers pillar page, the DOI entry, and the ROR ID registration guide for institutional identifiers. For the broader reproducibility policy context, see NIH’s rigor and reproducibility policy and the reproducibility dictionary entry.
