Messy but valuable research data risks sitting for eternity in silos unless efforts are stepped up to make it FAIR
Voices calling for wider access to, and reuse of, data from health and biomedical research are getting louder. Findability, accessibility, interoperability and reusability (aka FAIR) are crucial to allow analyses to be carried out on large-scale datasets in the drug development process. Recognising this, research gatekeepers, including the EU funding bodies, now mandate high quality data management for all projects.
OpenPHACTS, an early IMI project, was one of the ‘thought incubators’ cited in the drafting of the original FAIR data manifesto, published in 2016, with its OpenPHACTS Discovery Platform cited as an example of system in which FAIR principles were already being implemented. Its stated goal was to help drug discovery by cutting down on the time, effort and money required to find, connect and mine data from diverse private and public data sources, work that was wastefully repeated across companies, institutes and academic laboratories. This integration of public databases was done by companies themselves, a process that was repeated regularly when the databases were updated. OpenPHACTS did it once, publicly, in a manner that allows it to easily kept current, meaning companies no longer have to do it themselves.
The project did this by setting up an online platform that linked publicly available drug discovery databases to give researchers a one-stop-shop where they can quickly find the data they needed to test their scientific hypotheses. The platform covers diverse domains like proteins compounds, targets, diseases and tissues etc. and is freely available in the public domain. The platform was very innovative - one interface and data integration to allow for very complex searches all in one spot.
Plugged in
While the key strength of the platform is to make publicly available datasets available through a single, simple query, the pharmaceutical company partners in the project all had confidential data they wanted to integrate for their own use. OpenPHACTS, therefore built infrastructure to allow for this, while guaranteeing the confidentiality of the company data. This functionality has led to the founding of Phenaris a spin-out company that offers data, models, and decision support in all aspects of in silico toxicology.
Phenaris plugs toxicological data from eTOX, another early IMI project, into the OpenPHACTS infrastructure to allow researchers to predict candidate compound toxicity, and therefore decide which compounds to prioritise, saving time and money, and animal experiments. eTox broke ground in that it enabled pharmaceutical companies to share previously inaccessible proprietary data, giving a ‘unique, industry grade toxicology database’.