InChI Trust

InChI Trust is a United Kingdom nonprofit organization responsible for supporting the development, maintenance, and adoption of the International Chemical Identifier, commonly abbreviated as InChI and pronounced “IN-chee.” InChI is an open, non-proprietary textual notation for representing chemical substances from molecular structure data. The organization is best understood not as a conventional consumer or industrial brand, but as a standards steward and technical community institution serving chemistry databases, cheminformatics software, research organizations, publishers, and other users of machine-readable chemical information. The InChI project originated through work by the International Union of Pure and Applied Chemistry (IUPAC) and the United States National Institute of Standards and Technology (NIST) between 2000 and 2005. The first public release of the software and identifier appeared in 2005. The design sought to provide a structure-derived identifier that could support searching, indexing, and linking chemical information across databases and the web without requiring a central authority to assign registry numbers. Unlike proprietary or institutionally assigned identifiers such as CAS Registry Numbers, an InChI can be generated from a chemical structure using publicly available algorithms and is intended to be freely usable. An InChI encodes a substance through layered information. Depending on the selected form, those layers can describe the molecular formula, atom connectivity, hydrogen placement, charge and protonation, stereochemistry, isotopic composition, tautomeric information, and certain metal-related connectivity conventions. The algorithm processes an input structure through normalization, canonicalization, and serialization. Normalization reduces alternative representations to a common chemical interpretation; canonicalization assigns a reproducible numbering scheme; and serialization writes the result as a structured text string. This architecture allows chemically meaningful components to be compared across software and databases, although the standard does not represent three-dimensional atomic coordinates. A major related offering is the InChIKey, a fixed-length hashed representation designed for practical web searching. Full InChI strings can be long and difficult to use as search terms, whereas an InChIKey is compact and has a consistent 27-character form divided into sections. Because hashing is not reversible, an InChIKey must be connected to an underlying InChI through a database or resolver service. The InChI ecosystem therefore supports both identifier generation and interoperability among chemical information services. InChI Trust also supports extensions and associated technical formats, including Reaction InChI (RInChI), which represents chemical reactions through groups of InChI strings for reactants, products, and agents. The broader software ecosystem includes molecule editors, database pipelines, web services, standalone executables, and packaged implementations for operating systems such as Linux. These uses make InChI a foundational interoperability layer rather than an end-user product brand. Since May 2009, the InChI Trust has served as the nonprofit body through which the standard is developed and promoted. Its work includes coordinating technical improvements, supporting software releases, encouraging implementation by chemical-information tools, and preserving consistency across versions. Version 1.07.1, released in August 2024 according to the cited reference material, moved the software to the MIT license. This licensing change further emphasized the project’s open and broadly reusable character. The organization’s present role is consequently centered on maintaining a durable, community-oriented identifier standard for chemistry rather than selling chemical products or operating a conventional commercial portfolio.

The InChI Trust emerged from the development of the International Chemical Identifier, a project initiated by IUPAC and NIST in 2000. The project addressed a long-standing problem in chemical information management: chemical substances can be represented by systematic names, registry numbers, line notations, drawings, database records, and other formats, but these representations do not always provide a common, freely computable key for linking records across systems. The project’s objective was to create a textual identifier derived from molecular structure that could support database indexing and web search while remaining non-proprietary. Development continued through the first half of the 2000s, and the first InChI release was made in 2005. The identifier was designed as a layered representation rather than a simple name. Its main layer records the formula and connectivity of the normalized structure, while additional layers can encode charge, protonation, stereochemistry, isotope information, and other structural distinctions. The normalization process is important because different drawings or input conventions can otherwise produce inconsistent results. In particular, standard processing can reduce alternative tautomeric or protonation representations to a common parent interpretation. The project also created a software-based approach to canonicalization. After normalization, atoms are assigned a reproducible numbering and the resulting information is serialized into a defined string format. This makes it possible for independent software implementations to generate comparable identifiers from the same structural input. InChI is more structurally expressive than simple line notations in some respects, although it is not intended to encode three-dimensional coordinates. A practical limitation of the full identifier is its length. To support search engines and web services, the InChIKey specification was released in September 2007. The InChIKey is a compact hash-based form of an InChI. It is useful as a search and indexing token, but it cannot reconstruct the original identifier and is not mathematically guaranteed to be collision-free. Resolver services and chemical databases are therefore needed to associate keys with full InChI strings and molecular records. Version 1.02, released in January 2009, introduced the Standard InChI. This fixed the level of structural detail and conventions used for the standard form, helping different databases and software systems compare identifiers more consistently. In May 2009, the InChI Trust became the nonprofit organization supporting the continuing development and promotion of the standard. Its role included maintaining the software, coordinating technical work, encouraging implementation, and representing the identifier as an open infrastructure project for the chemical-information community. The InChI family subsequently expanded beyond individual substances. Reaction InChI, or RInChI, applies the underlying approach to chemical reactions by organizing identifiers for reactants, products, and agents. Auxiliary information generated alongside an InChI can preserve atom mappings, equivalence information, and data useful for reproducing or redrawing the input structure. These capabilities have supported integration with molecular editors, databases, resolvers, and research software. The project’s distribution model remained central to its identity. InChI was created as a non-proprietary standard that could be generated from structural information rather than assigned by a single registry organization. According to the cited reference material, version 1.07.1 released in August 2024 uses the MIT license. This provides a permissive legal basis for continued adoption and implementation. InChI Trust’s history is therefore primarily a history of technical standardization, open software stewardship, and interoperability in chemical data rather than commercial brand expansion.

• 2024

  InChI version 1.07.1 adopts MIT licensing

  [1]International Chemical Identifieren.wikipedia.org

  The August 2024 software release uses the MIT license.

• 2009

  Standard InChI introduced

  [1]International Chemical Identifieren.wikipedia.org

  Version 1.02 provides a fixed-detail Standard InChI and corresponding Standard InChIKey conventions.

• 2009

  InChI Trust assumes stewardship

  [1]International Chemical Identifieren.wikipedia.org

  From May 2009, the United Kingdom nonprofit InChI Trust supports development and promotion of the standard.

• 2007

  InChIKey specification released

  [1]International Chemical Identifieren.wikipedia.org

  A condensed hashed identifier is introduced to make chemical-structure searching more practical on the web.

• 2005

  First InChI release

  [1]International Chemical Identifieren.wikipedia.org

  The first public version of the International Chemical Identifier is released.

• 2000

  InChI project begins

  [1]International Chemical Identifieren.wikipedia.org

  IUPAC and NIST begin developing a machine-readable, non-proprietary identifier for chemical substances.

An open-standard steward for chemical structure identification and cheminformatics interoperability.

• Chemical structure identifier · 2005

  International Chemical Identifier (InChI)

  InChI is a textual representation of a chemical substance generated from structural information. Its layered format can encode formula, connectivity, hydrogen placement, charge, protonation, stereochemistry, isotopes, and selected tautomeric or metal-connectivity information. The algorithm normalizes the input, assigns a canonical atom ordering, and serializes the result. It is designed for database indexing, record linkage, and web-based chemical searching.

• Hashed chemical identifier · 2007

  InChIKey

  InChIKey is a fixed-length, 27-character hash-derived representation of an InChI. It is intended to be easier to search, store, and exchange than a full InChI string. The key is not reversible and has a small theoretical possibility of collisions, so it must be linked to a full InChI or chemical record through a resolver or database.

• Standardized chemical identifier · 2009

  Standard InChI

  Standard InChI is a constrained form of InChI that uses fixed conventions and a defined level of structural detail. By reducing variation in how identifiers are generated, it helps databases and independent software systems compare chemical records more reliably. Its corresponding hashed form is the Standard InChIKey.

• Reaction identifier

  Reaction InChI (RInChI)

  RInChI extends the InChI approach to chemical reactions. It organizes InChI strings for reactants, products, and agents and adds reaction-level information needed to distinguish their roles. The format is intended to improve machine-readable exchange and linking of reaction data across cheminformatics systems.

• Cheminformatics software · 2005

  InChI software and auxiliary information

  The InChI software generates identifiers and can produce auxiliary information describing atom mappings, equivalence relationships, and data useful for reconstructing or redrawing an input structure. Implementations are available through chemical editors, standalone executables, web services, and operating-system packages.

• 2024

  Adopt the MIT license for InChI softwareStrategy

  Background

  The project continued to emphasize broad reuse and open implementation of its chemical-identification software.

  Action taken

  Version 1.07.1, released in August 2024, used the MIT license.

  Aftermath

  The release provided a permissive licensing framework for users and developers incorporating the software into chemical-information tools.

  Source

• 2009

  Create a nonprofit stewardship organizationStrategy

  Background

  The InChI project required continuing technical maintenance and promotion after its initial development by IUPAC and NIST.

  Action taken

  The InChI Trust began serving as the nonprofit organization supporting development and adoption of the standard from May 2009.

  Aftermath

  The Trust became the principal institutional steward of the InChI software and associated standards work.

  Source

1. 1International Chemical Identifieren.wikipedia.org
2. 2InChI Trustinchi-trust.org