NEW YORK (GenomeWeb) – PerkinElmer has unveiled Signals for Translational, a new cloud-based data management, aggregation, and analysis platform that it has developed to help researchers in pharmaceutical companies and other contexts combine and analyze data from multiple sources, and has begun testing it ahead of a full launch planned for the first half of next year.
Karen Madden, the head of PerkinElmer's informatics group, said that the company has secured several early adopters for the platform, and expects to add a few more, to more fully develop and refine the system before making it more widely available commercially. The first full release date is still being discussed and will depend on the exact set of features that PerkinElmer decides to include, she told GenomeWeb.
PerkinElmer is also still working out the commercialization details but it expects to offer Signals under a software-as-a-service model for some customers, she told GenomeWeb, and will likely charge a yet-to-be-determined subscription fee that will be based on the specific applications and tailored to customers' usage and storage needs. She said that the company could provide additional pricing details closer to the launch date.
PerkinElmer launched Signals for Translational to address a perceived need within the pharmaceutical and biotechnology industry for integrating and analyzing large quantities of useful research data stored in disparate databases and repositories. "The way that data is being generated right now from all the different instruments and technological advancements that are out there is creating the need for scientists to be able to get their hands on the right data, at the right time in order to be able to make the right decisions," Madden said.
"One of the challenges is that [when] scientists generate all this data, it goes into different parts of the organization … and often the people who really need to see the data … have trouble getting their hands on and interacting with it and exploring it," she added. "That's what Signals allows scientists to do."
The need for data aggregation is felt most acutely by researchers involved in translational research studies, Jens Hoefkens, director of research and strategic marketing at PerkinElmer, told GenomeWeb, in part because it's a relatively new component of pharma and biotech companies and is still finding its footing. As a result, researchers working in this context often struggle to find adequate tools that allow them combine the datasets they need for their projects. "Longer term, we see many more applications for Signals both in basic research and clinical applications but our first focus is on the translational scientists," Hoefkens said.
According to Hoefkens, Signals features a data analytics component built on the Tibco Spotfire platform that provides tools for aggregating and querying various kinds of structured and unstructured data from a broad range of experiments and assay types such as next-generation sequencing including DNA-seq and RNA-seq; variant calls; gene expression assays; and immunohistochemistry and imaging, among others. Madden also told GenomeWeb that the engine underlying the Geospiza platform, which the company acquired in 2011, will also be incorporated and integrated into Signals for Translational.
Researchers can integrate experimental and clinical data from public sources such as the Gene Expression Omnibus, as well as from private and proprietary databases. Data can also be uploaded to the system from users' internal pipelines and processes via PerkinElmer-developed application programming interfaces. The system uses semantic web technologies to combine and connect different kinds of data stored in different formats.
Also, the system is flexible enough to adapt to new data types generated by new assay technologies, Hoefkens said. This is one of the factors he said that sets Signals apart from its main competitor TranSmart, an open source data sharing and analytics platform. Compared to Signals, TranSmart requires a "substantial" amount of development work in order to add in support for a new assay or technology, he said. With Signals, a new assay can be added in a matter of minutes.
Furthermore, although TranSmart is integrated with Spotfire, PerkinElmer's existing alliance with Tibco around Spotfire results in a much tighter, more seamless integration with the Signals platform than is possible with TranSmart, he noted. It's also simpler to load data into Signals than it is with TranSmart. Lastly since the TranSmart platform is free, users don't have access to the kind of commercial-grade support that they would get with Signals. TranSmart may be free, Hoefkens noted, "but to turn it into an enterprise-ready system and platform that you want to deploy in your organization and make a critical part of your business, there's a lot of work that needs to be done."
Moreover, Signals is agnostic to the type of data that it's being used to aggregate but has the ability to combine it with relevant domain expertise required for specific use cases.
"It's through the combination with scientific domain knowledge that we can create an application that is not just a generic [data] warehouse but is actually something that really speaks to the translational scientist by using the same terms, concepts, and entities that they are used to working with on a daily basis," Hoefkens said, adding that this is one of the features that sets Signals apart from more general purpose data aggregation systems repurposed for translational research use.
In the context of translational research, for example, the system is able to understand and make connections between patients and their clinical information such as things like doctors' visits, samples collected, and tests performed, allowing researchers to run specific queries on aggregated data, he said.
Another important benefit of the system is that it is cloud-based, which is a boon for potential users with limited budgets and, more generally, alleviates the burden of running and maintaining software locally, Hoefkens said. Running Signals on the cloud also supports collaborative research both within and outside pharma and biotech companies that would otherwise be difficult to accomplish with existing on-premise systems, he added.
One of the early adopters of Signals is Albany Molecular Research (AMRI), which is currently using it in its internal projects at its Buffalo, New York office and working with PerkinElmer to develop custom workflows and analytics capabilities that support high-content phenotypic screening projects. AMRI already uses a number of instruments provided by PerkinElmer including its cell imaging and robotic equipment and that made the company a natural partner for AMRI's informatics needs, Rory Curtis, vice president and head of AMRI Buffalo, told GenomeWeb. The company is running Signals on a private cloud and it has integrated the software with its existing imaging and robotic automation infrastructure from PerkinElmer. It is using the combined platform to analyze large quantities of complex digital cell images.
In addition to the existing relationship between the two companies, a big draw for AMRI was Signals' rapid time to results, noted Christopher Conway, AMRI's senior vice president for discovery and development services, which will enable AMRI to serve its own customers faster and more efficiently.
Signals also vastly simplifies the task of aggregating data from multiple locations, Grant Carr, AMRI's senior director of lead discovery, told GenomeWeb. While previously researchers might have had to search for information across a number of different sources and then manually integrate important information, Signals lets users visualize data points of interest across repositories via in intuitive interface, and offers statistical tools for analyzing and identifying important patterns in their data.
Although the initial target market for Signals is largely within the pharmaceutical and biotechnology industry, PerkinElmer sees it finding use in a much broader swath of the research market including in contract research organizations and the agbio industry, for example, where other PerkinElmer products are currently used, Madden said. The next set of applications planned for the platform is in areas related to translational medicine, starting with high-content imaging capabilities based on the work that PerkinElmer is currently doing with AMRI, she said.
The company is also looking to support imaging activities more broadly including in vivo imaging and quantitative pathology, she said, as well as specific clinical applications focusing on clinical trial data and potentially outcomes data related to clinical trials.