Syntensor

Syntensor is a life sciences AI company building biological foundation models to predict drug response in clinical trials and in individual patients. The platform converts large biological datasets into actionable decisions, enabling informed decision-making for investors, drug developers, regulators, and clinicians. Launched out of Cambridge University's entrepreneurial ecosystem, Syntensor is currently in beta and is developing some of the most advanced predictive tools in computational drug discovery.

Nearly 90% of drug candidates fail in clinical trials, most commonly due to insufficient clinical efficacy or unforeseen toxicity. Preclinical assays frequently fail to translate to the clinic, resulting in lost time, wasted resources, and unmet patient needs. Syntensor's platform addresses this by harvesting signal from assays across the drug discovery process, learning which data points translate to clinical outcomes and which do not.

Clinical Trial Outcome Prediction

• Predicts clinical trial outcomes from a drug's molecular structure with greater than 80% accuracy for recent Phase II trials and greater than 85% accuracy for Phase III trials
• Integrates data from multiple stages of the drug discovery process to identify predictive signals
• Designed to support decision-making earlier in the drug development pipeline, reducing costly late-stage failures
• Serves drug developers, investors, regulators, and clinicians seeking to de-risk development programs

Individual Drug Response and Genomic Foundation Model

• Expanding platform capabilities to predict drug efficacy and toxicity at the individual patient level, not just at the population level in clinical trials
• Patients with the same condition respond differently to drugs due to genetic and environmental factors; Syntensor's models aim to translate individual genomic variation into clinically actionable decisions
• Building a genomic foundation model trained on hundreds of thousands of individual human genomes to understand the architecture and diversity of the human genome
• Expanding on prior foundational work, including HyenaDNA, a genomic sequence model
• Targeting prediction of individual drug response as well as risk for complex diseases such as type 2 diabetes
• Designed to complement declining genome sequencing costs by providing interpretable, clinically relevant outputs

Technology and Research Foundation

• Applies biological foundation models and machine learning to drug discovery and precision medicine
• Research contributors include experts from Stanford, MILA (Yoshua Bengio's lab), MIT, Harvard Medical School, and CERN
• Incorporates geometric machine learning, graph neural networks (GNNs), generative models, and neural differential equations
• Core team includes ML researchers, bioinformaticians, drug discovery scientists, and software engineers
• Team members have prior experience at BenevolentAI, AstraZeneca, Novartis, Sumitovant Biopharma, and GLG

Team and Leadership

• Founded by Clayton Rabideau (CEO/CTO), whose background spans genetics, computational synthetic biology, and algorithm development
• Rosie Higgins (Cofounder and Advisor) brings experience as VP at Novartis and Data42, and COO at BenevolentAI
• Gabrielle Griffin, PhD (VP, Bioinformatics and Product) previously served as Director of Bioinformatics at Sumitovant Biopharma and technical lead for BenevolentAI's AstraZeneca partnership
• Callum Birch-Sykes, PhD (Senior ML Researcher) applies geometric ML expertise previously developed in particle physics research at CERN
• Craig Cinquina, VP Business Development, holds a PhD in Pharmacology from Yale and has prior experience at GLG and explainable AI companies Aitia and PredxBio
• Additional contributors include researchers from the Arc Institute, MILA, and Stanford working on explainability, long-context modeling, and generative models

Syntensor is currently offering early access through a beta community program, positioning itself as a transformative tool for both drug development decision-making and the emerging field of precision prescriptions.