Information technology began stealing the show in the drug discovery laboratory shortly after the decoding of the human genome at the start of the 21st century. It already had a pretty good head start, but once the fire hose of data from genome sequencing opened up, the hardware and software required for capturing, storing, and processing all that data couldn’t help but dominate.
During the past 15 years, the Bio-IT World Conference & Expo, held annually in Boston, has become a showcase of software and IT services for handling biological data. It also opens a window into the lab through which the progress of genomics and other data-intensive disciplines can be gauged.
As such, the event, which took place April 21–23, amounts to a reality check on the hype that tends to surround lab software, data storage, and the ever-expanding computing “cloud.”
The products highlighted by many vendors at this year’s conference link data storage products to services that help analyze that data. And there was an evident shift in focus by both vendors and other attendees from the promise of IT tools to the researcher the tools are meant to serve.
Michael Elliott, chief executive officer of the pharmaceutical IT consulting firm Atrium Research, pointed to evidence that drug research IT is emerging from the “hype cycle.”
“I think now you have a greater awareness and greater interest” in bioinformatics, he said. “Companies are making investments, looking at personalized or precision medicine, trying to aggregate data from the clinic, and looking at a lot of different services around translational medicine.”
In Elliot’s view, Bio-IT World brought forward evidence that technology support from the software industry is ramping up to accommodate the reams of data generated by next-generation sequencing. New tools and services for discovering and analyzing data are being offered as part of data storage systems, or warehouses.
“It’s an interesting trend, and there are a lot of little companies” offering the tools, Elliot said, “such as Content Analyst and Paradigm4, as well as some more established firms like Linguamatics and Cambridge Semantics.”
Ben Szekely, vice president of services at Cambridge Semantics, agreed that the research community is demonstrating growing interest in software that helps navigate data-inundated R&D.
“There is a greater acceptance from IT departments that you need an approach more flexible and scalable than a traditional data warehouse,” Szekely said of customer companies. “That’s the biggest change this year. People are seeking out these options rather than vendors having to jam it down their throats.”
Vendors, meanwhile, are adapting new approaches to data integration and visualization, he said. The result is new ways for researchers to query highly diverse types of data including genomic, phenotypic, and clinical data.
During the past two years, Cambridge Semantics has enhanced the chemical structure library in its Anzo Smart Data Integration platform, a data management and analytics system, with software supplied by ChemAxon, a Hungarian company exhibiting at Bio-IT World.
“ChemAxon brings the ability to index and search chemical structure information,” Szekely said. “We add that capability to our platform as a software component.”
ChemAxon’s chemical structure software has also caught the attention of ID Business Solutions (IDBS), a laboratory information management software company exhibiting at the conference. IDBS announced a partnership with ChemAxon through which the Hungarian firm’s software will be available with IDBS’s E-Workbook electronic laboratory notebook and ActivityBase data-screening platform.
Other deals were announced in Boston, including one under which Edico Genome, the maker of a computer chip that accelerates the speed of genomics data processing, will combine Intel’s Xeon processors with its Dragen field-programmable gate array computer chip. And DNAnexus, a bioinformatics infrastructure-as-a-service firm, announced a $15 million deal with WuXi PharmaTech in which it will provide cloud-based research services in China.
DNAnexus is targeting the genomics bioinformatics market by providing a cloud-based service that allows researchers to forgo investment in hardware or software for analysis of genomics data once it is produced by a next-generation sequencer.
“We are almost like the App Store at Apple,” said Brad Sitko, vice president of corporate development at the company. Users come with application requirements for genomic data analysis, and DNAnexus configures a platform for them on the cloud. “Our service is really the quality of our scientific staff, which can go toe-to-toe with any of the leaders in genomics.”
According to Sitko, researchers are getting over an initial hesitancy to commit to using cloud services. “A couple of years ago the concept of the cloud in genomics was a bit of a four-letter word,” he said. Lab managers are now sold on the benefits of cloud technology, he added, but require support in accessing off-site computing, especially given concerns about regulatory compliance, quality control, and security.
Support services and the ability to bundle applications for storage and data analysis are emerging as the keys to vendors’ success. Chris Sander, head of the computational biology program at Memorial Sloan Kettering Cancer Center, who delivered a plenary address at the conference, said pure technology offerings tend to be doomed.
“People come up with a technical solution and think they can form a company based on that,” he told C&EN. “But many of those companies have come and gone over the last 10 years, because software as such doesn’t sell. This has to do with the problem of how to connect to the brain of your user community.”
Firms that focus on services, on the other hand, are beginning to gain ground by getting into the user’s head and catering IT to research inquiry. “They integrate data storage and analysis services,” he said. Sander likens the success of firms tying technology to services to the evolution at IBM, which in the 1990s began abandoning a strictly hardware and software business to emphasize IT services.
Sander added that science know-how on the part of IT vendors and users is key to ensuring that nothing is lost in a reliance on huge data systems in drug research. “There is a strong risk of a superficial reduction of information,” he said. The more competitive suppliers are the ones that are familiar with the detailed information needed to appropriately parse the data.
“The end user only sees the tip of the iceberg,” Sander pointed out. “There is a whole pyramid underneath that, and if you don’t cover the whole pyramid in a technically competent way, you will lose essential information.”
The importance of scientists’ involvement in IT system development and implementation might have been best illustrated by the tranSmart Foundation’s receipt of a Best-of-Show Award in the Informatics & Data Tools category for its tranSmart Platform 1.2, an open-source bioinformatics software tool initially developed by Johnson & Johnson scientists that has been adopted by the foundation’s members, consisting of major drug companies, software suppliers, and IT consulting firms.
Despite advances in IT to manage reams of data housed outside of the laboratory, the researcher remains in charge, argued Rudy Potenzone, vice president of marketing for tranSmart.
“Our system is there to help the humans use their brains,” he said. “We are not trying to mimic the thought process. We are giving you data, well organized, so you can use your brain to figure out relationships and help you come up with the next set of questions to ask.”