1910, an innovative AI-native biotechnology firm, has introduced its groundbreaking AI model, CANDID-CNSTM, which significantly enhances the prediction of blood-brain barrier (BBB) penetration for compounds that exceed the traditional Rule of 5 (bRo5). Published in the Journal of Chemical Information and Modeling on December 5, 2025, this model offers a novel approach to drug discovery, particularly for central nervous system (CNS) therapeutics.
CANDID-CNSTM achieves an impressive 87% area under the precision-recall curve (AUPRC) for bRo5 small molecules, outpacing Pfizer’s CNS Model Performance Optimization (MPO), which only achieves 56%. The model also distinguishes between CNS penetrant stereoisomers with a success rate of 68%, compared to Pfizer’s 50% accuracy, marking a significant advancement in neuroscience drug development.
The publication titled “CANDID-CNSTM: AI Unlocks Stereochemistry and Beyond Rule of 5 to Predict CNS Penetration of Small Molecules” presents CANDID-CNSTM as a pioneering solution to two major challenges in drug discovery: predicting BBB permeability and incorporating stereochemistry. The BBB serves as a formidable barrier, blocking approximately 100% of large molecules and over 98% of small molecules from entering the CNS. This has long rendered neuroscience one of the most complex areas within pharmaceutical research and development.
Breaking New Ground in Drug Discovery
Historically, most approved CNS drugs have adhered to Lipinski’s Rule of 5, which limits molecular complexity. Conversely, bRo5 compounds, characterized by their larger size and intricate structures, remain largely unexplored due to challenges such as exclusion from standard medicinal chemistry designs and the inadequacy of existing computational methods to predict CNS penetration. Stereochemistry further complicates this landscape, as current models often overlook its impact on BBB permeability.
CANDID-CNSTM addresses these limitations head-on, effectively predicting BBB permeability for bRo5 molecules and understanding the stereochemical distinctions that influence CNS penetration. “Neuroscience has long been defined by what we can’t reach,” said Jen Asher, Ph.D., Founder and CEO of 1910. “CANDID-CNSTM expands the boundaries of what’s considered druggable in the brain. By overcoming the limitations of bRo5 design and learning stereochemical effects, it opens an entirely new bRo5 chemical space for CNS drug discovery – bringing us closer to effective treatments for diseases like Alzheimer’s, Parkinson’s, and ALS.”
CANDID-CNSTM employs an attentive graph neural network (GNN) architecture, surpassing existing models by accurately predicting CNS penetrant bRo5 molecules. The model demonstrates a 90% area under the receiver operating characteristic curve (AUROC) when selecting CNS penetrant molecules from 1910’s proprietary database, compared to 81% from traditional approaches.
Impact on Drug Development and Future Research
The model has already contributed to the discovery of 1910-102, a non-opioid, covalent small molecule inhibitor designed for chronic pain management. This initiative is partially funded by the National Institute of Neurological Disorders and Stroke (NINDS) through the Helping to End Addiction Long-term (HEAL) Initiative.
“CANDID-CNSTM does not just classify molecules – it recovers the physicochemical principles that drive BBB transport,” said Jesse Collins, Ph.D., Senior AI Research Scientist at 1910 and lead author of the publication. “Its predictions correlate with quantum mechanical hydration free energy, indicating that the model implicitly learns the thermodynamic determinants of passive permeability. That mechanistic signal enables CANDID-CNSTM to generalize and identify brain penetrant bRo5 molecules and stereoisomers.”
CANDID-CNSTM is part of approximately 100 AI models within 1910’s Integrated Therapeutics Optimization and Transformation Model (ITOTM) platform. This innovative platform combines extensive multimodal data, cutting-edge AI models, and high-throughput laboratory automation to streamline the identification of novel disease targets and the design of both small and large molecule therapeutics more efficiently than conventional methods.
As the only AI-native biotech dedicated to advancing drug discovery, 1910 continues to position itself at the forefront of this rapidly evolving field, utilizing AI to unlock new possibilities in therapeutic development. For more information on 1910’s pipeline and initiatives, visit their website.
