The Hyve is a 30 person open source bioinformatics services company from Utrecht, Netherlands, and Cambridge, MA, USA. DNAdigest invited them to write a blog post on the summit “Open Source Technologies for Precision Medicine” that they organised in the beginning of June 2015 together with a life science consultancy Proventa International.
The summit “Open Source Technologies for Precision Medicine” took place in London on June 03 and had good attendance from both industry and academia. The round table discussions and the panel discussion lead by Keith Elliston (CEO, tranSMART Foundation), John Wise (Executive Director, Pistoia Alliance), Paul Avillach (Assistant Professor, Harvard Medical School), Jay Bergeron (Director Translational & Bioinformatics, Pfizer), Gerrit Meijer (Professor, Netherlands Cancer Institute) and Kees van Bochove (CEO, The Hyve) resulted in some constructive conclusions about the current situation with open source as a means to achieve precision medicine.
The main conclusions from the meeting:
- The ultimate business driver for adopting open source technologies such as tranSMART and cBioPortal seems to be access to data. For pharma IT, this especially means easy access to data from academics and non-profits, as well as annotated public studies and studies from (public-private) collaborations. Of course there is also the obvious advantage of zero license costs and sharing of the software maintenance costs (e.g. adaptations of the software to new omics data types).
- At the moment, TranSMART is enjoying huge interest from many early adopters. Companies such as Pfizer and Sanofi, which already have a global rollout of the system, are focusing on bringing more internal studies into the system, as well as improving the analytical capabilities and connectors. A multitude of PPPs (eTRIKS, TraIT, Translocation etc.), non-profits and patient organisations (Michael J. Fox, One Mind for Research, Movember etc) is adopting tranSMART as a data (exchange) repository.
- One of the main questions around the business model of using open source software, is who is paying for the curation and annotation of public data. We had an interesting case in the meeting with COSMIC, a public resource from the Sanger Institute which recently switched their licensing structure from freely available to free for academic use only. The director of COSMIC was present at the meeting and was able to get some feedback from the group. As the annotation and curation burden grows, it is increasingly more difficult to get these efforts funded publicly.
- There is an interesting synergy between tranSMART and cBioPortal, the cancer genomics research portal from Memorial Sloan Kettering Cancer Center which certainly was the audience’s favourite at the meeting. One of the differences between tranSMART and cBioPortal is that where tranSMART is a warehouse for a wide range of data, cBioPortal focuses on alteration events (e.g. over-/underexpression, mutations, gene fusions etc.) for various omics data types, accompanied by intuitive analysis reports that help the user explore these specific alteration events in relation to clinical data.
At The Hyve, we are currently working on a number of cBioPortal extension projects, some of them – with large academic cancer centers. If other institutes and pharma companies would be interested in looking at cBioPortal for internal usage, either on top of tranSMART or standalone, we would certainly be happy to facilitate that.
Agenda, slides and other details of the meeting can be found here.