To establish requirements for extending the applicability of the IUPAC International Chemical Identifier to organometallic and coordination compound structures, so that developers contracted to the InChI Trust can enhance the InChI software accordingly.
> Link to InChI subcommittee
The IUPAC International Chemical Identifier (InChI) algorithm is now well established as a powerful means of denoting the chemical structure of a well-defined, small (
The IUPAC Division VIII InChI Subcommittee is now working on widening the range of applicability of the Identifier, and with this in mind is creating a number of projects designed to facilitate various extensions. In each case the aim is to define precisely the requirements for structural representation in areas of chemistry not currently handled satisfactorily by InChI, bearing in mind that these requirements are to be translated into extensions of the InChI algorithm.
Discussions will take place by e-mail and at two face-to-face (annual) meetings. Once established, the requirements will be applied to the necessary software development by contractors appointed and paid by the InChI Trust, a non-profit entity funded by external organizations that use and benefit from the InChI algorithm. The present project is devoted to representations of organometallic and coordination compound structures. Such structures are particularly difficult to handle because standard representations based on simple connection tables are not a good approximation to reality and different, conflicting, conventions are used by different databases. The project will examine the various types of representation in general use, and propose appropriate translations to forms that an InChI-type algorithm can handle. Test structures will be provided by Elsevier (10,000) and the US National Cancer Institute (250,000), and the members of the Editorial Board and International Advisory Board of Dalton Transactions will be consulted for advice on currently acceptable representations.
July 2015 update: The requirements statement is almost complete and includes over a hundred structures that will be more than detailed enough for the developer to work on. Over the last month, the chair has held a series of meetings with Ian Bruno (Cambridge Crystallographic Data Centre,CCDC) in order to discuss in more detail CCDC’s requirements. In addition to the V2000 and V3000 files prepared last year, Heike Nau (Reaxys) has supplied a set of several dozen images and V2000 connection tables that cover the chemistry found in the Reaxys database that we would hope that the InChI code will cover. CCDC have found analogous structures in their database that cover the same chemistry and are currently preparing connection tables.
Dec 2017 update: We have held two meetings in 2017 as part of the two broader InChI workshops in Hinxton and Washington. This has uncovered important disagreements between people who are familiar with the code as to the best way to proceed. We are looking into a mechanism for finding people to make a proper investigation into disconnection and normalisation of coordination and organometallic structures, potentially using the CSD APIs as a source of structures.
Page last updated 28 Dec 2017