Compendia, a bioinformatics firm that got its start last year by licensing commercialization rights to the University of Michigan’s Oncomine cancer gene-expression database, celebrated its first birthday this week with a bit of a coming-out party.
While the firm has remained relatively quiet since its launch, that’s about to change in the year ahead as Compendia ramps up its sales and marketing and product-development efforts, CEO John Freshley told BioInform this week.
The company, which employed three staff members in September, now has a headcount of 16, including a full management team of veterans from companies like Spotfire, Gene Logic, and Celera (see this week’s People in the News for details on the company’s executive team).
The rapid growth was supported by $1.8 million in debt financing that Compendia recently raised from the State of Michigan’s 21st Century Jobs Fund and Madison Capital. In addition, the company has signed 10 customers for the commercial version of Oncomine, called Oncomine Professional.
Compendia did not disclose the identity of its customers, but said in a statement that they include “8 of the top 20 oncology drug companies.” Freshley said that the company is in “active discussions” with another 15 potential clients.
Oncomine Professional, Compendia’s flagship product, is an integrated database of publicly available cancer gene-expression data — the world’s largest, according to the company, with data from 264 studies and 18,740 microarray experiments.
Oncomine was originally developed for academic use by Compendia co-founders Arul Chinnaiyan and Daniel Rhodes at the University of Michigan, and continues to be freely available to academic users (via www.oncomine.org).
Compendia was founded last year to develop a commercial version of the database after Chinnaiyan and his colleagues noted that interest from industry in commercial licenses was picking up [BioInform 03-17-06].
Freshley told BioInform this week that the company has successfully signed on a number of clients who were previously paying the University of Michigan technology-transfer office for access to the database.
“A couple of our existing customers have come through the academic model, which is in itself quite an achievement because they were paying the University of Michigan something, but they weren’t paying the University of Michigan licenses that would support a company and support additional investment in the technology,” Freshley said.
Freshley declined to provide pricing information for Oncomine, but noted that because Compendia uses publicly available data its subscription price is considerably lower than other commercial gene-expression databases, which can range in the millions of dollars.
“We invest a lot in the challenge of gathering public domain data, but it does not compare to the investment that Gene Logic, for example, expends generating data running Affy chips. It’s on a different order of magnitude,” he said. “Therefore, we also don’t need to charge our customers the same kinds of dollars in order to recoup our investment.”
While acknowledging that proprietary data has value, Freshley noted that it’s been difficult for bioinformatics content companies to build a successful business around that model.
“I think what has happened with a lot of database companies is they’ve been able to generate some short-term excitement around some novel content, but they haven’t been able to sustain that excitement because of the high price tag. It just gets scrutinized more closely every year come budget time, but unfortunately the costs of doing business don’t go down.”
Freshley stressed, however, that Compendia isn’t simply repackaging publicly available data. He said that the company integrates, curates, and annotates the information in Oncomine, and also provides visualization and data-mining tools to help end-user biologists easily interpret the data.
In addition, the company has just added a new capability that enables customers to analyze their proprietary data alongside the information in Oncomine in a secure environment.
“We’re really a software service company rather than a content provider,” Freshley said. “We just have a whole lot of content that makes the software much more interesting.”
More Data and New Products
Now that Compendia is fully staffed, it is embarking on an aggressive product-development strategy. Since launching last year, the company has more than doubled the size of Oncomine to more than 580 million datapoints, and Freshley said Compendia plans to continue doubling the size of the database every year.
In addition, the company is currently working on adding different types of genomic data to the resource in addition to gene-expression data. “We have made a commitment to our customers that we will be releasing our first array-CGH dataset this year,” Freshley said. “We have also started gathering microRNA data and some high-throughput proteomics studies.”
“We’re really a software service company rather than a content provider. We just have a whole lot of content that makes the software much more interesting.”
He added that the company is currently developing analytical methods to integrate those different data types and “to come up with an appropriate way to provide those results in the context of all the gene expression data that we’ve already gathered.”
The company is also preparing to release a new product called Oncomine Concept Map, which organizes the information in the Oncomine database based on gene combinations that characterize “molecular concepts,” such as disease subpopulations, drug treatments, pathways, transcriptional regulations, and protein-protein interactions. Researchers can navigate the information in the database via these concepts, rather than querying by genes or disease type, in order to quickly find sets of genes that are associated with biological phenomena.
Freshley said that Oncomine Concept Map was also developed at the Chinnaiyan lab at the University of Michigan, and noted that it builds upon similar efforts at Duke University [BioInform 06-28-04] and the Broad Institute [BioInform 09-29-06] to navigate large gene expression data sets via gene, chemical, or clinical “signatures.”
The Concept Map is currently in beta testing and will be available for Oncomine customers before the end of the quarter, Freshley said.
In addition, Compendia has set its sights on moving beyond oncology into other therapeutic areas. Freshley said that the company is partnering with several academic labs in order to confirm that its approach can work in other disease areas.
“Gene expression and oncology go well together,” Freshley said. “There’s a little bit of proof that there’s going to be as much interest in other disease areas as there is in oncology, so we’re going through that process.”
Cancer has certainly been one of the biggest areas of gene-expression research, and the amount of data generated in the field is orders of magnitude higher than most other therapeutic areas — a fact that Freshley acknowledged. However, he noted, “there is data available, and a lot of it is going to depend on having the right academic relationships.”
One way the company plans on fostering its relationships in the academic community is by taking over maintenance of the free academic version of Oncomine, which has remained under the auspices of the University of Michigan for the last year. “Compendia is going to be taking on that responsibility on behalf of the University of Michigan, and really adopting the academic community as our customer, and providing them with an increased level of support and capability than they’ve had in the past,” he said.
Freshley stressed that the company will continue to keep the database free for academic use. “Any data that we gather from the public domain will go back to the academic research community in a more usable form than we received it,” he said. “We may add other features on top of that that at some point we may charge for, but we will always give them back what we received.”
One of the company’s challenges in the year ahead, he noted, “is going to be to communicate that.”
Freshley declined to provide details on new therapeutic areas that the company is exploring, but Chinnaiyan told BioInform last year that the company was considering moving into cardiovascular and neurodegenerative disease.