Pistoia Alliance provides unified data model update years in the making

Published: 25-Feb-2020

Supported by Biovia, Elsevier, GSK, Novartis, and Roche, Version 6 of the Unified Data Model file format has been released, a significant update since V5 in 2018

The Pistoia Alliance, a global, not-for-profit alliance that works to lower barriers to innovation, has released its latest Unified Data Model (UDM) file format, supported by funding from Biovia, Elsevier, GSK, Novartis, and Roche.

Building on Version 5 from November 2018, Version 6 delivers a stable foundation for the collection and exchange of reaction data and empowers the community to move forward with its adoption.  The release is built upon standard (Allotrope-compatible) vocabularies for units of measure and analytical methods, allowing the systems to be interoperable.

As increased adoption of AI and ML in R&D continues, such industry-wide data standards are essential in the move towards the digitally driven Lab of the Future (LoTF).

Dr Steve Arlington, President of The Pistoia Alliance, said: “Following this latest update, the project is now shifting into a support phase to help companies adopt the UDM model. Most recently we have secured additional funding from BMS to allow us to continue the project to benefit users, and we are pleased to welcome Cynthia Beno from BMS to our steering committee.”

Arlington explained that the company is also currently looking for additional funding in order to develop a toolkit which would bring together methods, techniques and literature to support the adoption of the UDM.

The UDM

The latest update of the UDM looks to primarily improve the semantics and validation of data. This is important in R&D when acronyms and words can be interchanged. Without alignment on terms and data models, great opportunities to see trends or wider patterns are missed.

Standardised harmonisation of data is especially important as increased adoption of AI, and machine and deep learning is being seen, which are reliant on high quality data to improve outcomes and predictions. To continue to improve alignment in the future, the Alliance is looking to work more closely with the Allotrope Foundation, to ensure the two systems are complementary and compatible.

To help with the adoption of the UDM, Jarek Tomczak will also be presenting on V6 at the upcoming AI for Reaction Outcome and Synthetic Route Prediction conference.

“It’s fantastic to see the UDM continue to develop under the project team’s stewardship, and to serve the R&D ecosystem. Elsevier donated the original UDM to the Pistoia Alliance and we hope it continues to succeed and promote collaboration in the future,” said Tim Hoctor, VP of Professional Services at Elsevier and board member of The Pistoia Alliance.

Following the success of the UDM project, The Pistoia Alliance is currently developing a new project, SEED, which is complementary, and will provide guidance on the specifics of connecting a semantically enriched ELN to the FAIR principles, as well as open standards built to enable semantic tools to seamlessly connect to all ELNs. Looking to the future, the UDM project team will continue to support the UDM and look to develop the adoption toolkit.

You may also like