1	The IUPAC InChI project – A Status Report Stephen Heller InChI-Trust Project Director steve@inchi-trust.org
2	"The slides from this presentation..." The slides from this presentation can be found at: http://www.hellers.com/steve/pub-talks/(San Francisco 2010 link) and at the new InChI Trust web site The main web sites for the IUPAC InChI project are: http://www.iupac.org/inchi and http://www.inchi-trust.org
3	Disclaimer These slides were made from 100% recycled electrons. This will be a well balanced presentation. I have a chip on both shoulders. I will try to be politically correct, and have even take a course in being PC. But, I flunked it – twice.
4	Overview
5	Objective The objective of the IUPAC Chemical Identifier Project is to create a unique label, the IUPAC Chemical Identifier (InChI), which will be an Open Source, freely available, non-proprietary Identifier for well defined chemical substances that can be used in printed and electronic data sources thus enabling easier LINKING of and working with diverse data and information compilations.
6	Why Use InChI? For publishers and database providers using InChI gives one a competitive advantage being able to LINK content from multiple sources. It offers users the ability to help in new discoveries from existing information and data by easily being able to integrate, remix, and retell. Business models that depend on things not changing (closed access and making aggregation and integration difficult) are so 20^th century. InChI is a small, but vital, part of new business models and technologies involving chemicals that will lead to new discoveries. Combinability increases the value of information and data.
7	""In my view" "In my view, the most important rule of business in today's integrated and digitized global market, where knowledge and innovation tools are so widely distributed. It's this: Whatever can be done, will be done. The only question is will it be done by you or to you. Just don't think it won’t be done." By Thomas L. Friedman Published: December 9, 2008 NY Times (That is – not if, but when and by whom. The web has made the impossible to now be possible. )
8	I skate to where chemistry is going to be, not where it has been. With apologies to ice hockey great, Wayne Gretzky.
9	The Internet has made the world more homogenous for chemical information and the Open Source InChI/InChIKey is not affected by global boundaries or proprietary chemical structure representations
10	InChI is an agent of change for those individuals and organizations who have defined chemical structures and want to make their information known and available to the community.
11	InChI Technology Other Technology
12	Critical factors for the success of InChI project Technically competent staff Fulfill a real community need Political and Financial Support
13	Technical: InChI is a unique representation/identifier for defined chemical structures. Probably marginally better than previous ones. The InChI algorithm was built on the shoulders of giants. http://en.wikipedia.org/wiki/Graph_theory Practical: InChI and the related hash-code compressed InChIKey are the only available universal LINKs for in-house and public databases of defined chemical structures.
14	InChI is the worst computer readable structure representation except for all those other forms that have been tried from time to time. With apologies to Sir Winston Churchill (House of Commons speech on Nov. 11, 1947 )
15	Initial InChI Goal (Plan A) – Cover 100% of all chemical found in the literature and in databases and other sources. Current InChI Goal (Plan B) - Cover 99.9% of chemicals found in the literature and in databases and other sources.
16	Bar Codes – not designed to be read by humans InChI – not designed to be read by humans
17	Why InChI is becoming a success 1. Organizations need a structure representation for their content (databases, journals, chemicals for sale, products, and so on) so that their content can be LINKED to and combined with other content on the Internet. 2. InChI is a public domain algorithm that anyone, anywhere can freely use.
18	How do we know the InChI project is beneficial? Success is uncoerced adoption
19	InChI have some advantages over other chemical identifiers developed before: (1) They are freely useable and non-proprietary. (2) They allow a more advanced representation of chemical information than other codes (such as the SMILES code). (3) They are unambiguous, i.e. conversion of chemical structures using standardized algorithms only leads to one InChI. (4) They are precisely indexed by major search engines such as Google. However, InChI are not applicable to generic formats often disclosed in patent literature, such as Markush structures, since they were rather designed to represent specific chemical structures and compounds. InChI therefore are not yet useful for comprehensive retrieval of patent literature. Excerpt taken from: Full-text prior art and chemical structure searching in e-journals and on the internet – A patent information professional’s perspective World Patent Information, Volume 31, Issue 4, December 2009, Pages 278-284 Maik Annies
20	The best way to represent a chemical compound is not by a name or even a database identifier, but by its structure encoded in Structure Data Format (SDF MDL V2000) or the open Chemical Markup Language (CML) format or InChI codes. A few databases already provide the IUPAC/NIST standard of InChI codes or the shorter hashed InChIKey. The new InChIKey resolver services implemented by the Royal Society of Chemistry (RSC) and Chemspider allows to create InChIKeys from molecular structures and a reverse lookup of InChIKeys to obtain the associated known structures from molecular databases. The InChIKey can be used for web based literature search and also for chemical database search and merging of compound lists from multiple sources. Some other databases support the SMILES code for structures. The use of SMILES code is not recommended because multiple vendors create different representations of the SMILES code. Also true canonical (unique) SMILES are vendor specific. Extracted from: Kind T, Scholz M, Fiehn O: How large is the metabolome? A critical analysis of data exchange practices in chemistry. PLoS One 2009, 4:e5440.
21	The idea is to create a mechanism that would allow CrossRef publishers to record InChIs in their submitted CrossRef metadata. This, in turn, would allow us to provide a service that would allow users to: Lookup the published articles that mention a particular InChI. Lookup the InChIs mentioned in a published article. … The following is a demonstrator of what an DOI2InChI lookup service might look like. Please note that the XML representation of the results is very basic and is not best-practice for linked-data. The demonstrator currently only holds DOIs and InChIs for a few publishers. A summary of the contents of the database can be found on the status page http://inchi.crossref.org/status A list of all the CrossRef DOIs that contain InChIs can be seen here: http://inchi.crossref.org/dois A list of all the InChIs that have been registered with CrossRef can be seen here: http://inchi.crossref.org/inchis The system provides the following API calls: Return all the DOIs that have been registered with a given InChI http://inchi.crossref.org/dois/InChI=1S/C4H6O2/c1-3-6-4(2)5/h3H,1H2,2H3 Return all the InChIs that have been registered for a given DOI http://inchi.crossref.org/inchis/10.1038/nchem.215 © 2009 CrossRef (CrossRef Labs web site for CrossRef experimenting R&D)
22
23	The LINKED and Interoperable and Combinable World of InChI
24	InChI Trust Organization
25
26	InChI Trust Membership With the needs of NIST fulfilled with respect to what capabilities of an InChI are required for NIST databases, and since IUPAC is fundamentally and culturally a volunteer organization, there needs to be a way to continue development of InChI, and maintain the InChI algorithm. As a result of numerous meetings, emails, and discussions, it was concluded that a not-for-profit organization would best fit the project needs. Thus the decision to create and incorporate the "InChI Trust" in the UK. As there is no "free lunch", the Trust will need resources to continue to operate. Membership in the InChI Trust requires annual dues. The income from these revenues will be used exclusively for InChI development, maintenance, and educational activities associated with the project. Membership will entitle a member to influence the direction, priority, and speed of further Trust activities. Membership will also provide InChI Trust "certification" of the InChIs and InChIKeys in a member's database. Those organizations which do not join the InChI Trust will still have free access to the InChI algorithms but will not participate in any decision-making or direction -setting activities.
27	Summary If you are not part of the solution; you are part of the precipitate.
28	Current InChI Trust Members ACD Labs ChemAxon Dialog Elsevier FIZ CHEMIE – Berlin Informa/Taylor & Francis IUPAC John Wiley & Sons Microsoft Nature Publishing Group OpenEye Royal Society of Chemistry (RSC) Symyx Thomson-Reuters 3/7/2010
29	Acknowledgements (Primarily members for the IUPAC InChI subcommittee and associated InChI working groups) Steve Bachrach, Colin Batchelor, John Barnard ,Evan Bolton, Steve Boyer, Steve Bryant, Szabolcs Csepregi ,Rene Deplanque, Nicko Goncharoff, Jonathan Goodman, Guenter Grethe, Richard Hartshorn, Jaroslav Kahovec , Richard Kidd, Hans Kraut, Alexander Lawson , Peter Linstrom, Bill Milne, Gerry Moss, Peter Murray-Rust, Heike Nau , Marc Nicklaus, Carmen Nitsche, Matthias Nolte , Igor Pletnev, Josep Prous, Peter Murray-Rust, Hinnerk Rey, Ulrich Roessler, Roger Schenck , Martin Schmidt, Steve Stein, Peter Shepherd, Markus Sitzmann ,Chris Steinbeck, Keith Taylor, Dmitrii Tchekhovskoi, Bill Town, Wendy Warr, Jason Wilde, Tony Williams, Andrey Yerin. Special Acknowledgement: Ted Becker& Alan McNaught for their vision and leadership of the future of IUPAC nomenclature.