1910 Genetics Launches CANDID-CNS™, an AI Model for Blood-Brain Barrier Permeability Prediction
1910 Genetics, a biotechnology company specializing in small and large molecule drug discovery, has announced the launch of CANDID-CNS™, an AI model designed to predict blood-brain barrier (BBB) permeability and distribution of small molecule drug candidates targeting central nervous system (CNS) diseases. The model boasts an impressive 83% success rate in predicting brain penetration and distribution, significantly outperforming the industry standard CNS Multiparameter Optimization (CNS-MPO) score, which has a success rate of only 64%.
Given that only about 2% of small molecule drugs can cross the BBB, developing therapeutics for CNS diseases poses significant challenges. CANDID-CNS™ is part of 1910’s Multi-AI Agentic System within its Input-Transform-Output (ITO™) platform, which aims to revolutionize drug discovery for both small and large molecules.
CANDID-CNS™ utilizes an attentive graph neural network to enhance prediction accuracy and has been validated through a study published as a preprint titled “Attentive Graph Neural Network Models for the Prediction of Blood Brain Barrier Permeability” (DOI: 10.1101/2024.10.12.617907). This platform includes hundreds of proprietary models, facilitating the drug discovery process from hit discovery to lead optimization.
“Our results indicate that CANDID-CNS™ sets a new standard for predicting small molecule drugs that can penetrate the brain, achieving an 83% success rate compared to the traditional 64%,” said Jesse Collins, PhD, lead author and AI Research Scientist at 1910 Genetics. “The unique anatomy of the BBB complicates CNS drug development, making AI-driven solutions like CANDID-CNS™ crucial in identifying promising drug candidates.”
Jen Nwankwo, PhD, Founder and CEO of 1910 Genetics, emphasized the model’s significance for neuroscience drug development. “CANDID-CNS™ effectively addresses challenges in data scarcity and model training, maintaining state-of-the-art performance across multiple metrics, including logBB and Kp,uu.”
1910 Genetics offers its ITO™ platform to pharmaceutical and biotechnology companies through co-discovery, co-engineering, and platform-as-a-service partnership models. More information can be found at 1910genetics.com/partnering.
About the ITO™ Platform
1910 Genetics’ ITO™ platform is an advanced, scalable AI solution aimed at improving drug candidate discovery across various modalities and therapeutic areas. It consists of a multi-AI agent system with hundreds of models leveraging federated learning and three proprietary data streams: computational data, wet lab ground truth biological data, and wet lab proxy biological data. The platform empowers drug discovery teams to enhance target identification, molecule design, and ADMET property simulation, increasing the likelihood of clinical success while reducing drug development costs.
About 1910 Genetics
Founded in 2021, 1910 Genetics is a pioneering biotechnology firm advancing drug discovery with a multimodal AI platform enhanced by laboratory automation. The company integrates AI with three proprietary data streams to deliver innovative drug candidates and software solutions to pharmaceutical partners while advancing its pipeline in neurological, autoimmune diseases, and cancer. Recently, 1910 announced a partnership with Microsoft to scale its ITO™ platform for pharmaceutical applications.
With initial funding of $26.1M from prominent investors, including M12 (Microsoft’s Venture Fund), Playground Global, and OpenAI’s Sam Altman, 1910 Genetics has established its headquarters and laboratory in Boston’s Seaport District and continues to expand its expert team. For more information, visit 1910genetics.com.