IMI projects generate a lot of data, particularly the ones that deal in translational research (those that ‘translate’ basic science into medical practice). Such projects depend on knowledge management (KM) capabilities that allow researchers to work with primary study data (generated themselves) and secondary study data (generated by others). Lacking a common KM platform, IMI projects were investing in individual solutions, causing wasteful and redundant overhead costs, while also risking the legacy of the data they generate. Though there are many KM platforms on the market, commercial solutions don’t suit the requirements of research projects that are publicly funded, that have multiple partners, or that wish to allow open access to datasets.
The consortium delivered a knowledge management software platform called eTRIKS that is now available to IMI and non-IMI researchers. They created and deployed a host of analytics applications known as eTRIKS Labs, as well as a variety of best practices, guides and standards and other documentation. The consortium’s assets are mostly available under open licenses and the application of best practices continues through the work of the eTRIKS commercial spinoff, Information Technology for Translational Medicine (ITTM), the eTRIKS Data Sciences Network and the many adopters of eTRIKS’ technologies.
eTRIKS supported over 60 projects by the end of the collaboration (with an initial goal to support 40). At the final meeting in September 2019, it was noted that several additional research groups had utilised eTRIKS’ software and best practices, bringing the known project engagements up to 70. (eTRIKS personnel have not formally tracked use of the eTRIKS project assets since the end of 2017, although some insights were gained throughout the eTRIKS extension through to end 2018 and until the project ended.)
The platform
In order to make an ‘open’ platform, eTRIKS translational research information platform is based on tranSMART, an open-source data warehouse designed to store large amounts of clinical data from clinical trials and basic research. eTRIKS released five major platform versions during the course of the project. The consortium split the work into different streams that covered hosting, software development, deployment, analytics, data curation, mapping, testing, as well as data standards and ethical data use.
The consortium also deployed and hosted the publicly-accessible open access eTRIKS Public Platform which houses about 200 curated clinical studies from a wide variety of disease areas. Those interested in using the eTRIKS platform were invited to use the public platform as a testing and training environment and were provided with a variety of support documents for helping to use the platform.
Etriks Labs
The project resulted in a number of modular applications and tools that could be used to perform data analysis, exploitation and visualisation. eTRIKS Labs arose out of a desire to brand the applications and make them available to the wider research community. All eTRIKS Labs are open license. Sample apps/tools include eTRIKS Analysis Environment (eAE), a high-performance compute grid scheduler; SmartR, which provides a dynamic and interactive way of visualising and analysing data within tranSMART; SNF, a novel computational method for genomic data integration; and Patient Input Platform, a discussion game framework to assist patients and legislators in navigating the risks and benefits of consenting their individual health data.
Standards Starter Pack
Data standards are the rules by which data are described and recorded. In order to share, exchange, and understand data, consistent standards are essential. The Standards Starter Pack was developed by the consortium to document best practices for optimising the quality and usability of data loaded to the eTRIKS platform. Intended for project leaders and data managers, it provides a review of biomedical data standards as well as guidelines on which standards platforms are best suited for specific research plans. The Standards Starter Pack documents were made available for all IMI projects to promote consistency in data handling and to raise awareness of the advantages of applying consistent standards across translational research projects.