SAN DIEGO--(BUSINESS WIRE)--Iambic Therapeutics, a clinical-stage biotechnology company developing novel therapeutics using its unique AI-driven discovery platform, today announced Enchant™, a multi-modal transformer model designed to provide predictive insights into clinical properties of potential medicines from the earliest stages of drug discovery.
Enchant provides Iambic’s scientists a new in silico tool to reduce clinical risk by predicting the viability of molecules from the earliest stage of drug discovery programs. These predictions include pharmacokinetic (PK) properties – how the body interacts with drugs – enabling researchers to make informed decisions to boost a program’s chance of clinical success before entering the clinic. In turn, this could reduce the cost of drug research and development, as well as lessen the burden on participants in clinical trials.
“Bringing a drug to market often costs billions, partly because critical pharmacological insights aren’t uncovered until human trials are well underway,” said Tom Miller, PhD, Iambic’s Chief Executive Officer. “Enchant is designed to cut years from preclinical development, speed up trial timelines, and prevent late-stage discoveries of liabilities that can jeopardize clinical success.”
Enchant is a multi-modal transformer capable of processing multiple types of data simultaneously. Trained on dozens of public and private data sources and modalities, Enchant processes small amounts of human trial data in combination with abundant pre-clinical discovery data to better predict key drug properties that impact clinical outcomes.
“Enchant represents a new frontier in AI-driven drug discovery where deep machine learning can better predict how molecules will act and react in humans before they enter a clinical trial,” said Connor Coley, PhD, Associate Professor, Massachusetts Institute of Technology. “By leveraging Enchant’s predictive capabilities, developers can accelerate the process of designing, making, and testing drug candidates, ultimately reducing the burden of traditional experimentation methods.”
A series of demonstrations showing Enchant’s predictive power were published in a white paper released today:
- Even when trained on less than 1% of the Obach human PK dataset – representing training on data for just 5 distinct molecules – Enchant demonstrates meaningful predictive power which improves with increasing amounts of widely available preclinical data.
- When training incorporates the full Obach human PK dataset, Enchant sets a new benchmark for prediction accuracy for human PK properties: a Spearman correlation coefficient of 0.74 for human PK half-life represents a major advance over the previous state-of-the-art of 0.58.
- Enchant’s predictive power across a range of laboratory and clinical tasks outperforms previous state-of-the-art models.
“Enchant addresses a fundamental data bottleneck in drug discovery, supplementing scarce clinical data with readily generated and abundant preclinical laboratory data,” said Fred Manby, PhD, Iambic’s Chief Technology Officer. “When trained on the full set of available human clinical PK data, Enchant’s clinical predictions surpass all state-of-the-art models. But most critically the model gets better at predicting clinical outcomes by training on more laboratory data.”
As a fully multi-modal transformer, Enchant builds on Iambic’s prior work to address two key data challenges: different data types created at different R&D stages and the mixed quality of public data sources, which Iambic researchers have addressed by a process to standardize, clean, and transform data into streams of small units, or tokens, suitable for model training.
About Iambic Therapeutics
Founded in 2019 and headquartered in San Diego, California, Iambic Therapeutics is disrupting the therapeutics landscape with its unique AI-driven drug-discovery platform. Iambic has assembled a world-class team that unites pioneering AI experts and experienced drug hunters with strong track records of success in delivering clinically validated therapeutics. The Iambic platform has been demonstrated to deliver high-quality, differentiated therapeutics to clinical stage with unprecedented speed and across multiple target classes and mechanisms of action. The Iambic team is advancing an internal pipeline of clinical assets to address urgent unmet patient needs. Learn more about the Iambic team, platform, and pipeline at iambic.ai.