November - December 2005
Oogling for Chemists
Chemists have often berated the online world for a lack of a straightforward search engine that allows them to search for chemical information quickly and easily. Pioneers such as Cambridge University's Peter Murray-Rust, who we profile elsewhere in this, the fiftieth, issue of Reactive Reports, along with colleagues including Imperial College's Henry Rzepa is taking the chemical net forward in great strides with systems such as INChI. The National Institutes of Health too with its biocentric PubChem small molecule initiative is providing scientists with rapid access to a growing database of chemical information.
Now, San Diego based eMolecules Inc has launched what one might consider to be the chemical equivalent of the Google search engine - "Chmoogle". The company describes Chmoogle as the world's leading free open-access chemistry search engine and its mission is to discover, curate, and index all of the public chemical information in the world, and make it available to the public. "The world's knowledge in chemistry is an invaluable resource", said Klaus Gubernator, eMolecule's Chief Executive Officer. "It lies dormant until it becomes searchable by every chemist. The language of chemistry is chemical structures. Chmoogle makes the world's chemistry searchable by structure. Just draw a molecule using your favorite structure drawing tool and hit Go!"
"Currently, there are any number of excellent text search engines," Craig James, Chmoogle's Chief Technology Officer told us. "If you want to know everything about 'Alexander Fleming' (inventor of penicillin), they can help you," he adds. However, if you draw the penicillin molecule with your favorite chemical editor, can you search for it, James asks, "With Chmoogle you can."
Before Chmoogle, there was no free Internet resource of this nature. It provides a genuine cheminformatics system that anyone could use to find information using a substructure search. A number of academic institutions have searchable databases, but they're usually focused on their particular field of science, and their search systems are often primarily for organizing only their data. Chmoogle's goal is to be the search system to index all of the world's publicly available chemical information.
Chmoogle distinguishes itself from the attempts of other search engines to find chemical information by being extremely fast, it also offers an appealing presentation of results, and high-quality chemical drawings. "We've made it fast by bringing together a team with unparalleled knowledge in Cheminformatics," James told Reactive Reports, adding that, "Our team has decades of experience with every aspect of cheminformatics, from the core of the database engines to the applications that use it." He adds that in order to build Chmoogle, the team, "started from scratch, put everything we knew together, and built a system from the ground up, specifically tailored to handle the massive number of molecules we expected, and designed for the unique query patterns of a free Web search system."
James explains further, "The scale and speed of Chmoogle is unlike anything that's come before. We had to start from scratch, build a new chemical database engine from the ground up, so that we could give users the response times they expect, handle one of the world's largest collections of molecules, and respond to the unique demands of the Web."
Chmoogle goes deeper than a superficial search of the latent chemical Web, as one would hope. It allows users to send queries, results and individual structures as links to their colleagues using email. This, the company says, will create an unparalleled collaborative environment for chemists worldwide. Chmoogle also provides "Chmoogle Free" code that users can embed into their own Web sites for direct access to Chmoogle, as well as hosted cheminformatics systems and full Web sites for chemical suppliers, pharmaceutical and other chemical industries.
James does not see a conflict between Chmoogle and efforts to use the InChI. "Since Chmoogle will be crawling the Web looking for information, it may provide a valuable adjunct to their service: High-performance searches, combined with cross references to other places that Chmoogle has found the same molecule (for example, commercial vendors who have the molecule for sale, or other academic sites such as Zinc or PubChem)."
There is also a huge difference between search and retrieval. "InChI is excellent at retrieval," James says. "If you know the exact molecule you want," he adds, "the InChI string is unique so you can go straight to the record of interest." Chmoogle carries out substructure searching, which James explains, is a much more difficult problem to solve. "Substructure searching delivers very useful information to the user since it retrieves the query molecule, if it exists, as well as related molecules," he told us.
Interestingly, Chmoogle works with the ChemSketch plugin from Reactive Reports' publisher ACD/Labs. "I've tried ChemSketch," says James, "and it's remarkably easy to use with Chmoogle. I like the way it is seamlessly integrated for one-button answers. It doesn't require Java or ActiveX controls, so it will work on any Windows PC."
By the way, if you're struggling with the pronunciation of Chmoogle, start by thinking champagne, apt for a celebratory issue!