Confirmant came on the scene with a bang just under a year ago, debuting a prototype of its Protein Atlas database at the Cambridge Healthtech Genome Tri-Conference in Santa Clara, Calif., with all the hoopla of a pre-bubble tech startup.
Following its splashy coming-out party, the Oxford GlycoSciences/Marconi joint venture stayed pretty quiet for the remainder of 2002. But that’s not to say the 20-person company hasn’t been busy. Confirmant has added 4,000 experimentally validated genes to the 7,000 that made up the Protein Atlas last year, has made a few slight refinements to its sales model, and has signed its first customer — “a prominent US academic institution” that had yet to give its approval to be named at press time last week.
The deal adds much-needed weight to Confirmant’s claims that a market still exists for value-added genomic data. The key differentiator between Confirmant’s data set and other genomics databases, according to CTO Jonathan Sheldon, is that “we’re starting with the protein and working backwards. We just go from the protein to the peptides that we map then to the genome, to the Golden Path.”
Using peptides generated by OGS, Confirmant has built an annotated gene database for its 11,000 genes without the “well-documented flaws” of gene-prediction algorithms and EST sets, Sheldon said. In addition, he said, because the proteins come from both normal and diseased samples, a wealth of experimental information, including details about sample information, cell line, and disease pathology, as well as information about splice variants, can be accessed about each gene in the database.
Sheldon said that the company has identified about 3,000 potentially novel genes that are not in Ensembl and has also found about 2,000 novel exons in known Ensembl genes. “The key thing being we’ve validated those exons with peptides, so we know they’re there,” Sheldon noted.
In addition to the new data, Confirmant has tweaked its data access model a bit. In addition to a three-year database subscription model, the company is now offering a fee-for-service model to allow customers to slice and dice the database into subsets of interest — information on particular sequences or sample information or disease type, for example. Pricing for this service is variable, said CEO David Palmer, and is expected to be attractive to smaller biotechs and academic groups that don’t have the budget to pony up the estimated £2 million ($3.2 million) for a full Protein Atlas subscription. Of course, Palmer added, the £2 million price tag is for the completed database of 30,000 or so human genes, so early subscribers are not yet paying that rate.
Palmer said the company would begin to target a new customer base for its data — chipmakers — midway through 2003.
The only other change is that of Confirmant’s original vision to deliver its data through an ASP model, which has been put on the back burner for now, Palmer said. “It may come around at some point, but right now we’re not seeing the market.”