Symphony of data in Vienna

Vienna, the city of the Wiener Philharmonic Orchestra who among other things give the world the famous New Year concert, was the scenery when the IMI (Innovative Medicines Initiative) Open PHACTS project went from being a project to become the Open PHACTS foundation.

Open PHACTS describe itself as follows on their web page:

“Open PHACTS began as an Innovative Medicines Initiative (IMI) project to build a platform for researchers to access and query publicly-available pharmacological data. Together we have built the Open PHACTS Discovery Platform by bringing together vast amounts of linked data in a single integrated, open infrastructure.

All the data brought together by Open PHACTS is free to access, to help researchers in drug discovery to find and use the information they need.”

The place for the meeting was very well suited to the ceremony. First of all Vienna University is the first academic institution to be a member of Open PHACTS and Vienna University also was the host of the meeting. In 2015, Vienna University celebrated 650 years, indicating that it is a good and stable partner to be involved with.

Secondly, as mentioned in the beginning of the article, Vienna is a good scenery due to the fact that it is a city of great music and many famous composers. Open PHACTS is very much about creating music out of data. By linking data from all over the web, making it interoperable and easy to access, Open PHACTS, like a musical director makes each individual data set play the same song, in the correct pace, bringing their special “sound” and character to the big data symphony.

Some of the data sets included in Open PHACTS are DrugBank, ChemSpider, ChEMBL, ChEBI, ConceptWiki, WikiPathways, Geneontology, UniProt and nextprot. Readers might ask themselves why the public part of VigiBase® (the WHO international database of suspected adverse drug reactions) is not one of the data sources. VigiBase would neatly cover also post marketing safety aspects. This is something that also the author of the article asks himself and one of the reasons to attend the meeting.

Day one of the meeting was very much about the history of Open PHACTS as an IMI project including lessons learned, what was good and what was not so good. The most difficult thing was managing the Open PHACTS community. Many members of the project disappeared when for example changing employment but interestingly many also re-appeared as part of their new employee proving that working with Open PHACTS was both interesting and stimulating on a personal basis. Also working with data, from sources that are constantly changing and evolving, is challenging but the linked data model and concepts used in Open PHACTS proved to be a good and robust choice to handle such challenges.

Another big challenge not only for Open PHACTS is how to handle old versions of the software/API (Application Programming Interface) being developed. As external systems starts to integrate the API old versions must be kept running until all those systems have changed to the new API. How long should such a transition period be and how do you identify what external systems are connected?

On the whole the Open PHACTS project has been very successful, proving that an IMI project can actually go from being a project to something that will be sustained also after the project is finished. Much is still to be proved of course or as Winston Churchill said – “Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning”. It will be very interesting to see how the Open PHACTS foundation will evolve now when it is on its own.

At the end of the first day there was a poster session where 12 posters where presented followed by coffee break and mingling among the posters including voting for the best poster. They all had in common that they utilized the Open PHACTS platform for drug discovery research in various ways. The winner of the poster competition was Emily Jamieson, University of Glasgow & Eli Lilly and Company with the poster “Variants, Genes and Targets: an investigative pipeline utilising Open PHACTS linked data to establish meaningful connections”.

The first day was wrapped up with a ceremony where the Open PHACTS flag was handed over from the project group to the newly constructed Open PHACTS foundation group represented by the three major pharma company members as seen in the picture below. After the ceremony all participants were invited to a dinner at the Augustinerkeller in central Vienna.

Stefan Senger (GlaxoSmithKlein), Marren Derek (Eli Lilly) and Herman van Vlijmen (Janssen Pharmaceutica) receives the Open PHACTS flag from Gerhard Ecker (University of Vienna) and Barend Mons (Leiden University) - © Samuel Erik Colombo | Optical Engineers

The second day was about the future. Many challenges as well as opportunities were listed. Herman van Vlijmen representing Janssen Pharmaceutica talked about how OpenPHACTS had affected a big pharma company. From Pharma IT, who had to accept new linked data sources being part of the IT infrastructure, to Pharma Scientists discovering a whole new world of datasets opening possibilities never before imaginable.

Nick Lynch, Open PHACTS Foundation, encouraged everyone to “Come and collaborate”. There are many ideas for improvements, addition of new data sources but also maintenance of what is already in place. Open PHACTS need more partners and wishes to expand the network to get the “snow ball rolling”.

Dean Allemang, Working Ontologist LTD, emphasised that what has been achieved in the Open PHACTS project, from software (APIs), data curation and linking to community building was really impressing. The most important thing though is the community - anyone can build software and curate data – building a successful community is what makes the difference.

On the whole it was two very interesting days. There are many interesting opportunities but also many challenges ahead. The biggest challenge is clearly how to make the Foundation survive without any external funding. The Foundation must get a critical mass of users that ship in money and resources to be able to keep maintaining, updating and adding new functionality as well as data to keep relevant in a quickly evolving environment. It might be hard to be as successful and long lived as the University of Vienna or the Wiener Philharmonic Orchestra but we must hope that the inspiring setting for the meeting will positively affect the chances of success and that the symphony of data will not go silent.