This week, the Linux Foundation, a non-profit consortium that supports the Linux operating system, announced that the publicly available version of Selventa's Biological Expression Language, OpenBEL, is now one of its collaborative projects.
For an annual $100,000 membership fee, the Linux Foundation will provide a collaborative and organizational framework for the OpenBEL project that includes access to "important services" as well as experience in open source project development and governance.
"It’s a great fit for us because [OpenBEL] is really Selventa's foray into open source software," Ted Slater, Selventa's chief technology officer for the OpenBEL consortium, told BioInform.
"The great thing about it from our perspective is that the Linux foundation has massive amounts of experience that we think will really help us out," Slater said. For example, they will provide operational support "that frees the rest of the people working on OpenBEL to do the science, which is really the important part of it," he said. The arrangement also makes OpenBEL visible to the sizable community that contributes to the Linux Foundation's supported projects, which could up the number of contributors to the language and the framework, Slater added.
For the Linux Foundation, OpenBEL represents its first steps into the life sciences market, Jim Zemlin, executive director at the Linux Foundation, told BioInform. In addition to sponsoring the Linux operating system, the organization, which set up shop in 2000, provides services and infrastructure such as web and information technology management, hosting capabilities, intellectual property management, and more for industry and academic groups that want to collaborate on open source software development projects.
"The goal of these collaborative projects is to take the best practices that we've developed over time with Linux and work with other technology sectors of industry to use that same framework for development … in their respective marketplaces," Zemlin said. Usually, organizations that avail themselves of the foundation's services are "looking for a place that can host the development effort that’s trustworthy in terms of not taking sides with any of the particular competitors who may be collaborating together," he said.
Linux Foundation has hosted collaborative software projects for companies like Google, Cisco, HP, IBM, Intel, and Samsung.
OpenBEL was "was particularly interesting to us," Zemlin said, because it was an opportunity to explore how to "take the collaborative DNA of Linux and extend that to these other adjacent technical industries but very different from nuts-and-bolts IT."
OpenBEL provides a standard language and framework for capturing, storing, and sharing life sciences content. It lets users integrate knowledge across different representational vocabularies and ontologies, and combine knowledge from disparate sources into centralized knowledge repositories. The combined knowledge can then be made available to a variety of decision support and analytical applications through a standardized set of computable networks and application programming interfaces.
OpenBEL was developed by Cambridge, Mass-based Selventa and used internally to study drug efficacy and toxicity, identify mechanisms for drug sensitivity and resistance, and explore disease networks in greater depth. The company released BEL as an open source software project last year after announcing plans in 2011 to co-develop a public version of BEL with Pfizer (BI 4/1/2011). Since then, it has caught the eyes of firms like Foundation Medicine who, seeing a potential role for OpenBEL in its efforts to create cancer genomic analysis capabilities, have joined the OpenBEL consortium and will help guide further development of the language and associated tools and content, Slater said.
"The great thing about BEL is it was specifically designed from the beginning to represent in a qualitative way the causal relationships that can exist between biological entities," he told BioInform. "It's really easy to take the information in [a] journal article [for instance] that’s present in natural language and convert it into BEL, which is very simple and human readable [and] also machine readable and you can create a very powerful graph knowledge structure" for things like visualization and querying.
OpenBEL has been used for research projects done by AstraZeneca, the Fraunhofer Institute, Harvard Medical School, Novartis, Pfizer, and University of California, San Diego, and others. Overall, it's been used in more than 80 commercial life science projects in the course of its decade-plus existence, according to Slater.
It currently "covers a lot of ground" however, there is still room for improvement, he said. For example, it is useful for representing information in the early stages of the drug development pipeline but it needs to be extended to cover the latter stages such as clinical trials, he said. Other development efforts might focus on creating information visualization tools and enabling different forms of reasoning.
"One of the things we think holds lots of promise for BEL is interoperability with the semantic web standards like RDF," he said. Currently, "we have some software available for people now to go from BEL which is very human readable to RDF which is not incredibly human readable and back again."
Moving forward, the Linux Foundation is interested in using its services to support more life science projects. "We think we have a very unique set of experience and skills that we can extend to those industries and we hope to see more and more efforts like OpenBEL," Zemlin said.
Pricing for the foundation's services and the kind of service provided varies depending on the size of the project as well as its specific requirements.