Skip to main content
v2026.11,986 entries · CC-BY 4.0

Chemical Database: Types, Major Examples and How to Search

Chemical databases by content type, with an aspirin identifier cross-walk, search steps and a task-to-resource table for researchers.

Written and maintained by CASRAI Editorial Board

Published

On this page

A chemical database is a searchable collection of records about chemical substances. Each record ties a substance to identifiers such as a name, a CAS Registry Number or an InChIKey, and to data such as structure, measured properties, bioactivity, regulatory status or the literature that reports it. No single database covers all of that, so the practical skill is knowing which kind of chemical database holds the answer you need and which identifier gets you to the right record.

What is a chemical database?

A chemical database stores one record per substance (or per reaction, assay or document) and lets you retrieve records by a key. The key is usually a name, a molecular formula, a structure drawing or an identifier. The CAS Registry Number (CAS RN) page describes the problem identifiers solve: compounds are described by molecular formulas, structures, and generic, systematic, common and trade names, and that lack of clarity can cause delays and even safety concerns. CAS describes a CAS RN as a unique and unambiguous identifier for a specific substance, and says CAS Registry Numbers have set the standard for identifying chemical substances since 1965.

That is why most chemical databases are really two things in one: a registry that decides which record a substance is, and a data layer attached to that record. When you compare databases, ask two separate questions: how does it identify a substance, and what data does it hang on that identity?

Types of chemical database by what they hold

The databases researchers use most fall into five groups. The groups overlap, and large resources such as PubChem span several of them, but the grouping predicts what a search will return.

Substance registries and identifier services

Registries assign or collect identifiers and link them to structures and names. CAS REGISTRY is the best-known example; CAS also runs CAS Common Chemistry, which it describes as an open community resource with nearly 500,000 chemical substances drawn from CAS REGISTRY. For uncommon substances the CAS page points to a lookup service rather than the open resource, and the full CAS data is licensed.

Evaluated property databases

These hold measured physical and chemical data that experts have reviewed. The NIST Chemistry WebBook (Standard Reference Data 69) is the standard example: it lets you search by name, formula, IUPAC identifier or CAS number and returns data such as thermochemistry, vibrational and electronic spectra and phase-change data, each credited to its compilers. Use it when you need a value you can cite as evaluated data, not a number scraped from a supplier page.

Bioactivity databases

Bioactivity databases connect compounds to their effects on targets. ChEMBL, hosted by EMBL-EBI, describes itself as a manually curated database of bioactive molecules with drug-like properties that brings together chemical, bioactivity and genomic data. For binding-affinity measurements specifically, see BindingDB and DrugBank, and the CASRAI dictionary entry for ChEMBL.

Regulatory and toxicology databases

These organise substances around use, exposure and hazard. The U.S. EPA CompTox Chemicals Dashboard lets you search by chemicals, by products and use categories, and by assay or gene. On 2026-11-19 its home page showed version 2.9.0 and 1,386,492 chemicals. The CASRAI guide to the CompTox Chemicals Dashboard covers its curation and batch-search workflow.

Reaction, patent and literature databases

Some databases are indexes of documents and reactions rather than of substances alone. CAS describes CAS SciFinder as a chemical compound database covering substance and reaction data, literature, patents, regulatory information and supplier data, searchable by structure, name, CAS Registry Number or spectra data. It is a licensed product, and its content varies with the license. The CASRAI guide to SciFinder and the comparison SciFinder vs Reaxys cover that group in depth.

Who runs the major databases and how open they are

Openness changes what you can do with the data, not just whether you can read it. The data table below compares five major chemical databases by steward, main content, search keys and access terms. Three points stand out from it.

  • Government and standards bodies publish the evaluated and regulatory sets. NIST maintains the WebBook and EPA maintains the CompTox Dashboard. NIST states a licensing statement for Standard Reference Data on each WebBook page.
  • Open and licensed versions of the same registry can coexist. CAS Common Chemistry is open, while CAS REGISTRY data is licensed through CAS.
  • Licence terms differ even among open resources. ChEMBL says its current release is provided under a Creative Commons Attribution-ShareAlike 3.0 licence, which carries attribution and share-alike conditions that matter if you redistribute derived data.

The sibling CASRAI guide on PubChem covers the NCBI repository that most researchers use as a first stop, and ChemSpider covers the Royal Society of Chemistry’s aggregator.

How to search a chemical database: identifier first

Names are the weakest key. The same substance can carry many valid names, and one name can point to several substances. Searching by an identifier, then confirming the structure, avoids most mismatches. A repeatable sequence:

  1. Start from the strongest identifier you have. A CAS RN or an InChIKey beats a common name. The NIST WebBook search offers Name, Formula, IUPAC identifier and CAS number as separate options.
  2. Resolve it to a record in an open registry. PubChem returns a compound record identifier (CID) for a name, a CID or an InChIKey through its PUG REST service.
  3. Check the structure, not only the name. Compare the molecular formula and the structure drawing against what your source (a paper, a safety data sheet, a supplier label) actually describes.
  4. Carry the InChIKey into the next database. It travels between systems because it is derived from the structure rather than assigned by a registry.
  5. Record which database and which identifier you used. This is what makes a data pull reproducible in a methods section or a data management plan.

Worked example: one substance, five keys

The table below follows aspirin across two databases. The values were read from the PubChem PUG REST service and from the NIST Chemistry WebBook aspirin page on 2026-11-19, and the identifiers agree where the two systems share them.

Key Value Read in
Common name aspirin PubChem name search; NIST page title
IUPAC name 2-acetyloxybenzoic acid PubChem
Molecular formula C9H8O4 PubChem and NIST
CAS Registry Number 50-78-2 NIST (also among PubChem’s synonyms)
InChIKey BSYNRYMUTXBXSQ-UHFFFAOYSA-N PubChem and NIST
PubChem CID 2244 PubChem (the InChIKey resolves to the same CID)
Molecular weight 180.16 (PubChem); 180.1574 (NIST) Both; the difference is rounding, not disagreement

The molecular-weight row shows a real hazard: two reputable databases print the same quantity at different precision. When you cite a value, cite the database and its precision rather than copying a rounded number between sources.

CAS RN, InChIKey and why they behave differently

The InChI Trust FAQ explains that the InChIKey is a fixed-length 27-character signature built from a hash of the InChI string, so it is short enough for search engines and databases to index. The same FAQ says the chance of two different compounds sharing an InChIKey is not zero but small, so a key match is strong evidence, not proof, and the structure check in step 3 still applies. It also states that InChI is free. A CAS RN is different in kind: it is assigned by CAS when a substance is registered, and, per CAS, the open Common Chemistry resource and the licensed REGISTRY data sit side by side.

Strengths and limits

Strengths

  • Identity resolution. Registries and open aggregators turn a messy name into a record you can cite.
  • Traceable curation. NIST credits named compilers for each evaluated data set, and ChEMBL describes its content as manually curated, which tells you how much review stands behind a value.
  • Cross-database portability. Open identifiers such as the InChIKey make it possible to move a substance between resources without retyping names.

Limits

  • Scope is narrower than the name suggests. The WebBook is property-focused, ChEMBL is bioactivity-focused and CompTox is regulatory and toxicological. A missing record in one is not evidence the substance is unstudied.
  • Licensed content is not portable. Full CAS data and SciFinder content depend on an institutional or individual licence, so a result you can see may not be one a collaborator can reproduce.
  • Precision and versions drift. The aspirin example shows precision differences, and every database updates. Note the version or date: the CompTox Dashboard reports a version number and the NIST WebBook gives a DOI for the collection.
  • Identifier collisions are rare but possible. Treat an InChIKey match as a lead to confirm against structure.

Choosing a database by task

The table below maps common research tasks to the type of resource that is designed for them, based on what each steward says its resource holds. It is a starting point, not a ranking, and it makes no claim that one resource is better overall.

Task Resource type to start with Why (per the steward) Access
Confirm what substance a CAS RN or name refers to CAS Common Chemistry or PubChem CAS offers Common Chemistry as an open resource with nearly 500,000 substances; PubChem resolves names, CIDs and InChIKeys Open
Find an evaluated thermochemical or spectral value to cite NIST Chemistry WebBook Evaluated data sets credited to named compilers; DOI for the collection Open, with an SRD licensing statement
Find bioactivity of a compound against targets ChEMBL (see also BindingDB, DrugBank) Manually curated bioactive molecules with drug-like properties CC BY-SA 3.0 for the current release
Check regulatory and toxicology context for a chemical EPA CompTox Chemicals Dashboard Searches by chemicals, products and use categories, and assay or gene Open (U.S. government site)
Search reactions, patents and literature for a substance CAS SciFinder (or Reaxys) Substance, reaction, literature and patent data in one platform Licensed

Researchers who want the field’s wider context can read the guide to what cheminformatics is, which covers molecular representations such as SMILES and InChI and how open chemistry data fit the FAIR principles.

Citing and reusing chemical database data

For reproducibility, record four things whenever a database value enters a paper, dataset or data management plan: the database name, the identifier you searched with, the record identifier you landed on, and the date you read it. Some resources make this easy. The NIST WebBook aspirin page declares its parent collection with the DOI https://doi.org/10.18434/T4D303 and carries a licensing statement for Standard Reference Data. The ChEMBL home page lists its database paper under doi 10.1093/nar/gkad1004 and states the licence of the current release.

When a collaborator is on a different licence, reuse is the question that fails first. Open resources with attribution or share-alike terms can still be reused, but the terms travel with the derived data. Licensed resources usually restrict redistribution, so keep the identifiers and the query, which you can share, separate from the licensed content, which you may not.

Frequently asked questions

What is a chemical database?

A chemical database is a searchable collection of records about substances, reactions or assays, retrievable by identifiers such as names, formulas, CAS Registry Numbers and InChIKeys. Some focus on identity (registries), some on measured properties (NIST WebBook), some on bioactivity (ChEMBL) and some on regulation and toxicology (EPA CompTox Dashboard).

What is the difference between a CAS Registry Number and an InChIKey?

A CAS Registry Number is a unique identifier assigned by CAS to a specific substance. An InChIKey is a 27-character hash of the structure-derived InChI string, generated by software rather than assigned by a registry. Both appear on the NIST WebBook aspirin page (50-78-2 and BSYNRYMUTXBXSQ-UHFFFAOYSA-N).

Among those described here, CAS Common Chemistry, the NIST Chemistry WebBook, the EPA CompTox Chemicals Dashboard, ChEMBL and PubChem can be searched without a licence, although each has its own reuse terms. CAS REGISTRY data and CAS SciFinder are licensed.

Can I search a chemical database by InChIKey?

Yes in PubChem, whose PUG REST service resolves an InChIKey to a compound record identifier. The same InChIKey for aspirin returns CID 2244.

Sources

Data: Major chemical databases compared by steward, content, search keys and access terms

Compares six chemical information resources by steward, main content, search keys and access terms. Compiled on 2026-11-19 from each steward's own home or product page, the CAS Registry page and the PubChem PUG REST service.

Period covered: 2026.

Major chemical databases compared by steward, content, search keys and access terms
DatabaseStewardMain contentSearch keysAccess terms
NIST Chemistry WebBook (SRD 69)NISTEvaluated physical and chemical property data credited to named compilersName, formula, IUPAC identifier, CAS numberOpen; SRD licensing statement on each page
CAS Common ChemistryCASOpen subset of CAS REGISTRY substances (nearly 500,000)CAS RN and nameOpen community resource
CAS REGISTRYCASUnique CAS Registry Numbers identifying specific substances since 1965CAS RNData licensed through CAS
PubChemNCBI, U.S. National Library of MedicineCompound records with identifiers and propertiesName, CID, InChIKey (PUG REST)Records readable through the public PUG REST service
ChEMBLEMBL-EBIManually curated bioactive molecules with drug-like properties; chemical, bioactivity and genomic dataSearch in ChEMBL (entity-based)CC BY-SA 3.0 for the current release
EPA CompTox Chemicals DashboardU.S. EPAChemicals, products and use categories, assay and gene data (1,386,492 chemicals on 2026-11-19)Chemicals, Products/Use Categories, Assay/GeneOpen U.S. government site

Sources

Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · full access

Ask about Chemical Database: Types, Major Examples and How to Search

Full access: large monthly usage, up to 10 source passages behind each answer, and MCP and API access for Claude Code, Codex and other CLIs. Cancel any time.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →