We're building the world's first neuro-symbolic AI platform to understand disease biology, helping scientists go from hypothesis to successful experiment in days, not years.
The Ultimate Frontier for AI: Decoding Disease Biology
In the world of drug discovery, every decision is a high-stakes gamble. Despite an explosion of scientific data, the pace of innovation has slowed. It can take over a decade for a new drug to reach the market, and over 90% of clinical trials fail. Why? The leading cause is a fundamental lack of understanding of the underlying disease biology.
This is where the real challenge lies
The most important scientific breakthroughs are also the hardest to find, buried under mountains of complex data. Without a deep, unified view of underlying disease mechanisms - from genetic modifiers to disease outcomes - R&D teams face blind spots, irreproducible findings, and fragmented data. This leads to billions of dollars lost in failed programs and, most importantly, delays in getting life-saving treatments to patients who desperately need them.
This "black box" of disease biology is the ultimate frontier for AI
AI offers an unprecedented opportunity to finally decode these complexities, revealing the hidden connections and insights that have historically been impossible to see. By applying AI to disease biology, we can transform the entire drug discovery process, enabling a new era of breakthroughs and accelerating the delivery of effective therapies.
Introducing ASCEND by BenchSci
A New Era for Unraveling Disease Biology
ASCEND is the first neuro-symbolic AI platform designed to transform how biopharma understands disease biology. At its core is a rigorously vetted Biological Evidence Knowledge Graph (BEKG), curated by industry experts to ensure all information is backed by experimental evidence. This unique BEKG integrates diverse scientific data, including publications, multi-omics, and clinical trials, to provide a unified view of diseases.
By combining the BEKG with advanced foundation models, ASCEND offers groundbreaking solutions like AI copilots and co-scientists that accelerate discovery. We can also ingest your internal data to create a proprietary, customized map of disease biology, ensuring a secure and tailored experience.
Unparalleled Data Access
BenchSci's platform leverages a unique and comprehensive data approach. Unlike platforms limited to public data, our knowledge graph integrates tens of millions of scientific publications, including crucial closed-access journals through direct agreements with major publishers. We also ingest a client's proprietary internal data, such as lab notebooks, creating a single, comprehensive knowledge base. This ability to combine licensed, public, and internal data provides unmatched information breadth.
Biological Evidence KG and Data Quality
The core of the platform is a Biological Evidence Knowledge Graph (BEKG), a unique "evidence backbone" that ensures data quality and mitigates AI hallucinations. The graph structures over 400 million entities and a billion relationships, with every piece of information linked directly to traceable experimental sources. This process is supported by a proprietary multimodal AI that extracts insights from text, experimental figures, and supplementary files. A dedicated team of over a hundred scientists provides human-in-the-loop curation, ensuring scientific validity and consistency, which differentiates it from generic models.
Deep Domain Expertise in Complex Biopharma Workflows
BenchSci distinguishes itself through a deep understanding of complex biopharma workflows. Our platform is not a separate tool but an integrated solution that supports and enhances how scientists already work. It thinks like a scientist, translating natural language queries into complex, multi-hop scientific reasoning that enables a deeper level of discovery.
Change Management, Deployment, and Broad Adoption
BenchSci has a proven track record of successful deployment and adoption, with thousands of scientists using our platform across leading pharmaceutical companies. We provide a dedicated team for each account, including PhD-level scientific liaisons, who work directly with your researchers. This team ensures the platform is integrated into existing workflows and that your teams are fully empowered to maximize its impact. By focusing on augmentation, not replacement, we overcome industry caution and accelerate R&D.
Investors
Our Advisory Board
Guiding BenchSci's vision is a powerhouse Board of Directors and Advisory Board — a collective of leaders in life science R&D, machine learning, and culture. They provide the critical direction and expertise needed to revolutionize how new medicines reach patients.
Mikael Dolsten
Former Chief Scientific Officer and President of Pfizer R&D
Matt Truppo
Global Head of Research Platforms for Sanofi
Philip Tagari
Former VP of Research (Therapeutic Discovery) at Amgen
Philip Larsen
Former SVP and Global Head of Research at Bayer Pharmaceuticals
Scott Oloff
SVP and Global Head of Data and Analytics at Roche Pharmaceutical
Steve Hitchcock
Former Chief Scientific Officer at Takeda Pharmaceuticals
Sylke Poehling
SVP and Global Head of Therapeutic Modalities at Roche Pharmaceutical
Jean-Michel Lemieux
Former CTO/VPE Shopify & Atlassian
Jessica Neal
Venture Partner at TCV
Richard Zemel
Professor of Computer Science at the University of Toronto
