The recent article "Moving Towards Making QSARs for Toxicity-Related Endpoints Findable, Accessible, Interoperable and Reusable (FAIR)" addresses a significant challenge in computational toxicology: how to ensure that QSAR models based on machine learning (ML) and artificial intelligence (AI) are not only developed but also made accessible and usable.
On the second day of the PARC Modelers' Meeting [1,2,3], on June 10, 2025, in the session "Making Data and Models FAIR", Dr. Uko Maran gave a presentation entitled " Applying the FAIR data principles to modelling/models – experience from the QsarDB perspective ".
As a warm-up for the PARC Modelers’ Meeting [1,2,3], the QsarDB team provided a seminar on June 9, 2025, to offer attendees an overview and explanation of the basics of the QsarDB.org smart repository. The presentations covered various aspects of how to implement open and FAIR approaches to represent and organize machine learning (ML) and artificial intelligence (AI) models, as well as their underlying data.
We proudly announce that the QsarDB repository was awarded the CoreTrustSeal certificate of trustworthy data repository.

Recently we have been asked whether the models stored in the repository are compatible with the OECD QSAR Toolbox. Some of them are, and this feature has been available and tested since 2018. Unfortunately, feature has been unannounced and not well known. To improve the documentation, we made a sho...