Dec 2005 / Jan 2006 | Looking for a hot informatics market? Would you settle for one that’s lukewarm but growing decidedly warmer? The market for pathway analysis tools is just such an animal, though it’s still far from being a vast market able to quickly catapult a company to riches. At least one pathway provider, GeneGo, says it is profitable now, and in October, the FDA licensed GeneGo’s platform, MetaCore, for use by FDA researchers reviewing genomic data.
Steady growth in literature citations plus several major pharmaceutical deals and now the recent FDA commitment suggest pathway technology is gaining sufficient credibility and the furthest along commercially of all the systems biology niches. Broadly, pathway companies combine data-mining techniques with public or proprietary databases to identify pathways of interest. A simple use case is one in which a researcher inputs a list of differentially expressed genes from an experiment and is able to identify pathways of interest far more quickly than slogging manually through literature.
In recent years, there’s been a marketing war between pathway companies that use manually curated, proprietary databases and those that employ natural language processing (NLP) techniques to crawl and mine public-domain data. Manual curation is more costly but also more accurate, say proponents. NLP is less expensive and more likely to include the absolute latest results, say its advocates.
No doubt each has its place. GeneGo has bet big on manual curation. About 60 of GeneGo’s 70 employees work on curation, says Julie Bryant, GeneGo’s VP of business development. Ingenuity Systems, often perceived as the pathway market leader — a contention hotly contested by GeneGo — also uses a proprietary, manually curated database. Ingenuity’s DB is mostly compiled by an army of contract scientists. Ariadne Genomics is an example of a company in the NLP camp.
Virtually all pathway companies are trying to make it easier for biopharma to incorporate internal research databases with those supplied by pathway providers.
Part of what sets GeneGo apart is the size and diversity of its database. MetaCore has more than 5 million findings. “We are also the only platform to offer merged metabolic and signaling pathways. We have the only metabolic parser in the world today that allows you to bring in metabolomic concentration data from mass spec and visualize that in pathways, and we can also bring SNP data as well,” says Bryant.
You can tell the competitive juices flow freely in this market. GeneGo was founded in 2000 by Tatiana Nikolskaya, who trained at Moscow State University and the Graduate School VNIIGenetika and was a research associate at both the Department of Molecular Genetics & Cell Biology and the director of the Viral Core Facility at the Department of Medicine at the University of Chicago.
It’s difficult to know who’s really ahead in this market, since most companies are private. Bristol-Myers Squibb, GlaxoSmithKline, and Johnson & Johnson are among GeneGo’s customers. Bryant insists GeneGo has more revenue, more customers, and a more flexible business model than any other pathway tools provider.
“We have the Web model access and then in-house installations. We have named user licenses and we have concurrent floating licenses. We have multiple terms so the subscription is three-year, five-year, or perpetual,” says Bryant.
The company selectively offers a second license, called MetaBase, in which “we pull up the hood and say here are the schema, here are all the tables. When they have that, they can suck out the data or they can put data in so they may say, OK this is your tissue database and we want to add our tissue database,” says Bryant. MetaDrug is a third GeneGo product aimed at toxicogenomics.
Given the broad capabilities of pathway finders, and the narrow focus of disease model-makers such as Entelos and Gene Network Sciences, it’s interesting to speculate that a merger between a modeler and pathway house might prove beneficial. Such speculation aside, the pathway tool piece of the systems biology community seems closest to widespread commercial success.