MANCHESTER, UK--Bioinformatics software vendors and users have until August 24 to decide if they want to propose common standards for sharable software objects used to analyze genetic sequences. That deadline was set by a formal request for proposals approved here in late March by a committee of the Object Management Group, a nonprofit computer industry consortium that promotes interoperable software based on Common Object Request Broker Architecture (CORBA).
The new request for proposals is the first major product of an eight-month-old effort to create a common Interface Definition Language (IDL) that will allow informatics researchers to share data and software easily (see BioInform, March 2, 1998). The effort has been especially supported by large pharmaceutical companies, which are leery of investing too much money in software that can't communicate. The request, posted on OMG's web site (http://ww.omg.org/library/schedule/Biomolecular_Sequ._Analysis_RFP.htm), represents the most organized attempt to date to forge any kind of IDL standard within the often chaotic informatics community.
"It's a big step," Eric Neumann of Netgenics, a Cleveland-based software company, told BioInform. "There are really high expectations that this is going to produce a major response from vendors and that vendors will work together to produce some very high quality submissions." Neumann is a member of OMG's Life Sciences Research Domain Task Force, which prepared the request.
The Biomolecular Sequence Analysis RFP, as the request is formally known, notes that "despite many efforts over the years, the ability to consistently and precisely represent sequence and sequence-related information in a standardized form has eluded the computational biology community. The absence of such standards hinders the development of large-scale software architectures for directed, high-throughput analysis of sequence data, as is required in many applications in pharmaceuti cal drug discovery, the Human Genome Sequencing Initiative, and basic research in comparative and functional genomics."
To solve that problem, OMG is asking companies and academic researchers to propose "specifications of services for the analysis of biological macromolecular sequence data and associated information, including facilities for their representation, manipulation, and analysis."
Specifically, the RFP solicits proposals for sequence analysis services addressing three questions:
* What do sequence analysis objects look like?
* What is the nature of their inputs and outputs?
* How do these objects interact?
The RFP does not, however, cover graphical user interfaces (GUI's) for the display and visualization of sequence data.
Overall, submitters should "address the standardization issue through the use of IDL interfaces, formal descriptions of the behaviors of objects expressed by those interfaces, and clear statements of the biological concepts represented by them," according to the RFP. "The interfaces solicited should describe components that can be assembled into larger-scale applications and accessed by GUI-based CORBA clients for display and visualization."
Companies and researchers that send a letter of intent to participate by August 24 will have until October 19 to submit final responses. After a lengthy review process, OMG's 800-plus members will vote on final standards, probably in mid 1999. Once the consortium adopts a specification, it is made available for use by both members and nonmembers.
Neumann said that requests for proposals in other areas of interest to the bioinformatics community will be forthcoming. The sequencing request "may not address the broadest need," he said, "but it is an immediate need that needs to be satisfied quickly."