What's Happening?
OpenADMET has announced a new challenge focused on Cytochrome P450 (CYP) inhibition, a critical aspect of drug metabolism. This initiative, supported by data from Octant and UCSF, aims to provide high-quality datasets to evaluate the predictive power
of ADMET models. The challenge targets four key CYP isoforms: CYP3A4, CYP2C9, CYP2D6, and CYP1A2, which are essential for small-molecule drug metabolism. Participants will predict direct-inhibition pIC50 values and classify time-dependent inhibitors. The challenge is designed to improve the understanding of drug-drug interactions and reduce late-stage drug development failures.
Why It's Important?
The challenge addresses a significant hurdle in drug development: unpredictable drug exposure due to CYP inhibition. By slowing the clearance of drugs, CYP inhibition can lead to toxic plasma concentrations, posing risks such as drug-drug interactions and narrow therapeutic index issues. This initiative aligns with FDA guidelines to evaluate these risks and aims to reduce the need for late-stage clinical studies. The challenge encourages the development of cleaner drug candidates and supports the FDA's New Approach Methodologies, which seek to minimize animal testing in preclinical development.
What's Next?
Submissions for the challenge will open in approximately two weeks, with the competition running on a dedicated platform. Participants will have access to extensive training data, and the challenge will be evaluated using metrics like the Matthews Correlation Coefficient for classification tasks. The initiative also includes an award for innovative machine learning approaches, encouraging novel solutions in computational methods. The challenge timeline extends to November 2026, with results and insights to be shared through webinars and blog posts.
