Genentech open sources Unison: [Via business|bytes|genes|molecules]
Image by Getty Images via Daylife While on the subject of open and pharma, a bioinform article (sub reqd) tells us about Unison, a protein sequence analysis platform from Genentech that has been released under the Academic Free License (why not the Apache License since they are very similar). What is Unison? Unison is a compendium of protein sequences and extensive precomputed predictions. Integration of these and other data within Unison enables holistic mining of sequences based on protein features, analysis of individual and sets of sequences, and refinement of hypotheses regarding the composition of protein families
Essentially Unison is a data warehouse, which includes a number of protein sequences, and a bunch of pre-computed data. They have also released the complete schema, API, and some of the predictions. The backend is PostgreSQL and the platform leverages the BioPerl API. So the web service serves as a reference implementation of the Unison platform. People can essentially replicate the system and contribute code within their own servers using.
I think that biotech/pharma companies may do this more and more. The advantages for a company do not really come from these particular tools but how they are used and interpreted. Making this available to a much larger group means it is more likely to yield useful results. Genentech can only do so much with these tools. If someone else uses them to find something novel, some thing that Genentech did not recognize at all, Genentech might be able to reap some rewards that they would not have if they had kept things to themselves. Even if they do not get rewards directly, the publicity is worth something. They see this as a way to extend their influence rather than something for competitors to use against them. By furthering collaboration and increasing the number of eyeballs using their tools, Genentech can accomplish some things that would be difficult to do with their cards held close.
Technorati Tags: Bioinformatics, Health, Science, Web 2.0