ProteomeCommons.org Tranche – A commons for proteomics data

At the recent US HUPO conference, I happened to walk by a poster entitled “ProteomeCommons.org Tranche: Publicly storing and sharing large amounts of proteomics data”. I ended up having a short chat with the developer of Tranche, Jayson Falkner and was quite impressed with the effort. What is Tranche? Let me quote the poster introduction.

The ProteomeCommons.org Tranche project is designed to allow publication of all raw data relating to proteomics experiments. The system allows investigators to securely share files with collaborators, reviewers of manuscripts, and to publicly disseminate their data. Encryption and digital signatures allow authentication of data files, so provenance and integrity are unambiguous. The system provides for distributed downloads, background file compression, and allows file replication across the network to minimize the chance of file loss. While the system is intended to solve the raw data distribution problem, it is independent of file type so it can be used for peak lists, search results, images, and any other form of digital information.

The idea is that when people publish papers using mass spectroscopic methods for proteomics, the raw data are rarely included. The Tranche project essentially provides a resource for researchers to upload their raw data, share it, make it available to referees, etc. But the developers have gone one step further than just providing a repository. The framework looks quite solid, allowing data replication, self hosting, network hosting, etc. According to Jason they started off by implementing a BitTorrent, but a lot of people were apprehensive of the negative connotations that BitTorrent generates, so they ended up developing their own framework.

The ProteomeCommons.org moniker is somewhat unfortunate, since the organization is not affiliated with Science Commons. However, there has been a dialog between the two organizations recently and there is a possibility that ProteomeCommons.org might adopt Creative Commons licensing. That certainly makes sense and hopefully will not meet any resistance from the user community.

There is a Google Group for the effort where you can get a copy of the HUPO poster (which has more implementation details). I was just pleased to note that there were people in the community who understood the role internet technology can play in creating community and of the value of data sharing and transparency. There seems to be a fair degree of adoption among users with servers at some of the top proteomics facilities around the world. Another database project, the Human Proteinpedia, from the Pandey Group targeted for storing annotations and metadata related to proteomics experiments also plans to (perhaps already is) using the Tranche project as a repository for all raw data.

Update: According to Jayson Falkner (see comments), ProteomeCommons.org is in the process of moving from Apache2 licensing to CC licensing.

Technorati Tags: ProteomeCommons, Tranche, Proteomics, Open Data