In a strategic shift that it has been preparing since the middle of last year, GenoLogics is moving from providing research informatics and lab-management software to becoming a translational research informatics software provider, a company official told BioInform this week.
Two weeks ago, the company disclosed its first comprehensive project in this area, a collaboration with the Fred Hutchinson Cancer Center to migrate its in-house biorepository and biomedical informatics software solution to the GenoLogics platform. This week, the company announced a co-development partnership with GenVault for the first module for its translational research product portfolio, a biospecimen data management system called BioVault.
Translational research informatics is a “new major focus for us,” James DeGreef, company co-founder and VP of market strategy, told BioInform. While the company’s research informatics and lab management software — such as Geneus for genomics data management, Proteus for proteomics data management, and the LIMS platform OMIX — remain core company products, GenoLogics decided to expand its scope to target a new market opportunity.
“Translational research is really driving the demand for our core products and our new products,” said DeGreef.
Biomarker discovery has intensified in recent years, particularly the validation of these markers, whether for early detection, targeted therapeutics, or other applications. With the need for larger cohorts and observational studies, findings need to be validated on a larger scale. This holds true for the pharma world, too, where biomarkers are used for efficacy or safety in drug development as well to stratify a population for targeted therapeutics and clinical trials.
To meet these demands, GenoLogics is developing a multi-module biomedical informatics suite: BioVault for specimen management; BioChronicle for clinical annotations management; BioQuest for Web access; BioSphere for study management; and BioSource for electronic patient questionnaires. Although the company has officially launched the product line, most of the modules are still in production.
“There are going to be a number of modules released this year, next year and more into the future,” DeGreef said.
BioVault is currently available in beta format. “We are working with early-access customers, generally customers who have our research informatics [software]. We are in major development right now,” he said.
As part of this new focus, GenoLogics has hired several new employees, mainly from Canada and the US, but also from Europe, Japan, and the UK. These hires include a product manager as well as software engineers, field application scientists, and customer solutions staff to manage the early access program and larger deals.
Since it was founded in 2001, GenoLogics has focused on research institutions. But over the last year, interest from biotech and pharma companies has increased, DeGreef said, and a number of new accounts will be announced in the future, though he did not provide further details.
Translating the Data Flow
Market feedback is part of what led GenoLogics to translational research. Customers “don’t want to have to build everything from the ground up and they don’t want to have to deal with as many different small vendors who have a hard time integrating it all,” said DeGreef.
DeGreef said that customers have provided positive feedback for GenoLogics’ LIMS platform, but many were also looking for the company to help “upstream in terms of biorespository management, integrating information from clinical sources to help drive that information into the research domain, [offer] ways to query and request specimens, manage IRB [and] patient consent, [and provide options for] observational study management.” That demand created the second focus for the company, he said.
This space is not void of competitors. For example, InforSense has recently accelerated its push into the translational research informatics area. Last month, the company announced a three-year collaboration with Dana-Farber Cancer Institute to develop a translational research informatics infrastructure [BioInform 04-18-08], and earlier this month it signed an agreement with LabVantage Solutions to integrate its software with the company’s Sapphire LIMS with the intention of initially targeting biobanking management [BioInform 05-02-08].
“Translational research is really driving the demand for our core products and our new products.”
But DeGreef said that GenoLogics has more of a focus on data integration and data management than its competitors, for example pulling data and clinical information together from the hospital and the clinical trials systems, managing biorepositories, and managing lab instrument data.
“Traditionally we have provided mechanisms to get the data out into SAS or Spotfire or some of the other analytics tools, but they are not really purposed for translational research,” he said.
DeGreef suggested that the company’s shift in focus might bring about new partnerships, for example with analytics firms or other third-party software providers.
“There are big LIMS vendors that play into many different domains, but they are not really targeted, and then there are some smaller software vendors that might have a very focused solution around one science domain, but they wouldn’t scale up,” DeGreef said, “so there is sort of a sweet spot around being enterprise-enabled but having all the deep science needed for systems biology, which I think is the other side of the coin to translational medicine.”
The company is also considering a move into analytics, such as building analytical data marts, providing data to statistical analysis packages, or integrating with biological modeling systems such as Ingenuity and GeneGo. “That is less of a product [area, and] more of a service [area] and working with our customers,” he said.
Another aspect that played into the shift toward translational research is the search for an additional growth strategy. In recent years, GenoLogics’ products have been used more for discovery work, he said, but a market survey by his company revealed that many groups are now performing validation studies on biomarkers using larger patient cohorts.
One customer, for example, is a biotech company he declined to name that uses Geneus to manage its genomics data but its main problem is finding good specimens focused on the disease area it is targeting and the clinical information around those specimens.
“That biotech wouldn’t buy our new biomedical informatics system but they would be a direct beneficiary of it because it would be housed in [a] hospital or … cancer center, and [the biotech] could go to a web portal, query what specimens are there, and request them,” he said.
Robustness is a big theme in translational research, and DeGreef noted that many labs have high-throughput instrumentation performing complex experiments. “You can [handle that data] on a small scale, piecemeal, without our software for sure, but with translational research and trying to validate biomarkers, there are so many confounding factors, you have to take a real enterprise approach to it,” he said.
Validation is Today’s Job
The Fred Hutchinson Cancer Research Center has selected GenoLogics to develop an integrated biorepository and biomedical informatics software solution for its Translational Outcomes Research group.
“That is an approach more devoted to principal investigators,” DeGreef said, as opposed to “more of an enterprise approach” that another, undisclosed, customer is taking to build its biomedical informatics infrastructure.
Organizations may have biobanks at different locations and DeGreef said that as translational becomes a management focus, these groups are looking for ways to integrate these repositories.
GenoLogic’s software will allow clinical and research staff to manage their specimens locally but use a web portal to integrate all the clinical information and provide “sort of a virtual biobank at the enterprise,” he said.
GenoLogics believes its products can relate well to changes in the research landscape. In many cases, biobanks have in-house developed systems, perhaps built on Access or Excel. But as research teams scale-up their efforts, sharing specimens across institutes, to validate results, for example there is a renewed interest in robust systems “to be able to collaborate and share data.” There is a need to use controlled vocabularies to assure consistency in collaborative efforts between several groups and to automate outside requests from external groups as well as manage the IRB-approval process
“The major driver around translational research is massive collaboration, being able to validate your findings on other people’s samples, so there is a major market requirement putting together more … robust solutions that will allow people to scale up what they are trying to do,” he said. In-house systems might work but may not scale well, he said.
DeGreef cited the National Institutes of Health’s Clinical and Translational Science Awards as a potential driver for translational research informatics tools. NIH launched the CTSA program in 2006, and this week awarded an additional $533 million over five years to 14 medical research institutes under the initiative.
Not only will the data amounts increase as a result of this new funding, but collaboration is built in to these grants, DeGreef said.
Migrating a Growing System
Over the last decade the Fred Hutchinson Cancer Research Center has developed its own informatics system, with a grant from the Department of Defense. “We learned of the plusses and minuses of doing that,” said Nicole Urban, who heads the Translational and Outcomes Research Group at the center.
That system was developed in partnership with Beth Karlan, who directs the Women's Cancer Research Institute at Cedars Sinai Medical Center. Her system predates the Hutchinson one and was developed by Oracle programmers. Karlan was going to donate it to the Hutch, “but we couldn’t afford it because we didn’t have Oracle programmers,” said Urban. “We looked at her system and altered the specifications to meet our needs.”
The problem with the “home-grown system” is that it is programmed in FoxPro, a language that that “seemed fine at the time but it probably won’t be supported very much longer,” Urban explained.
The fact that they have a “pretty good system” is perhaps one of the reasons GenoLogics was interested in collaborating with her cancer center, said Urban. “We actually have a comprehensive functioning system of the type that everybody wants but we needed to replace it,” she said. It is “quite cumbersome” and also lacks a good querying system, because the modules of the systems are not well-linked.
For example, pulling data from blood specimens from patients who had developed resistance to chemotherapy with a focus on women with a strong family history of cancer can be done quickly but requires a database manager.
“The family history data is coming from questionnaires, and mutation status is coming from questionnaires and clinical chart review and then there is follow-up data that is in another system and then the information about the specimens is [in another system],” she said.
“We have a lot of specimens and we get a lot of requests from people to use our specimens and just trying to query the database to see if we have what they need is not easy,” she said. Being able to let the investigators query the database themselves should help the center save the money it needs to pay for maintenance of the GenoLogics system, she said.
After considering another home-grown system, Urban and her colleagues realized they had staff limitations that would slow down the development process. So she and her team spoke to many vendors. “None of them seemed to quite get what we needed,” she said. But as it turns out, GenoLogics had been planning to build exactly the modules that the Fred Hutchinson Center had been planning. “They can also hire a team of ten people to build this system, which we would never have been able to do,” she said.
The matchmaking between GenoLogics and the Fred Hutchinson Center was set up through the Canary Foundation’s founder Don Listwin, a former Cisco executive who quit his job to help accelerate research. When he asked her about her needs, Urban replied she had a database management system that “was not going to last forever.” The result is a donation by the foundation of $150,000 to cover the upfront cost of the GenoLogics system, she said. The Hutch will bear the cost of the data migration and the maintenance fee of $30,000 a year.
Urban said that the Hutch would have preferred to work with an open source software vendor, which GenoLogics is not, so the maintenance contract with the firm is a bit of a concession. She said she is also unsure if the GenoLogics system will be flexible enough as the center grows.
However, she said that it so far appears that GenoLogics is building tools that will follow the development at Fred Hutchinson, such as adding new data elements to the dictionary. “Even though we don’t have access to the source code, we will have tools to continue to develop the system ourselves,” Urban said.
For now the system is being built according to the Fred Hutchinson Center’s specifications as well as those of other clients, “but I don’t think any of the other clients have asked for anything we haven’t asked for,” she said.
Of late, what is becoming increasingly clear, is that annotated specimens are a necessity, she explained. “Getting all that information into the database is part of the work, but getting the information back out of the database in an easy way is also important,” she said. That step involves quality control and heeding patient confidentiality. “We have to be careful that it is not possible to identify an individual based on the annotation.”
Next on the schedule is a data migration phase. That will probably be a year-long process, requiring a deep understanding of the data dictionary. “It’s a big deal and in a way it is the scariest part of it,” she said.
For now, the in-house and GenoLogics systems will be run in parallel. The center already has a backup system in place, a mirror database. “We will do the same thing with the GenoLogics system,” she said, adding that the systems will run concurrently “until we’re comfortable.”