Leash Biosciences

Leash Biosciences is an artificial intelligence (AI) and machine learning (ML)-native biotechnology company based in Salt Lake City, Utah. The company aims to revolutionize medicinal chemistry by developing a foundational and generalizable machine learning model capable of accurately predicting small molecule drug candidates for any protein in silico. Additionally, Leash seeks to model interactions between proteins and chemicals.

To achieve its goals, Leash is compiling massive datasets comprising protein targets binding to chemicals. As of April 2024, the company had physically generated over 17 billion high-quality protein-chemical interaction measurements. Leash plans to expand this dataset by screening more than 500 protein targets against millions of proprietary, machine learning-designed chemicals by 2025.

Leash's approach involves harnessing the power of AI and ML, leveraging large-scale datasets to drive advancements in medicinal chemistry, akin to the transformative impact of data curation on fields like chess, image recognition, and protein folding. The company's machine learning engine will support the progression of multiple internal therapeutic programs toward in vivo studies.

Key customers and partnerships

In April 2024, Leash Biosciences launched the Big Encoded Library for Chemical Assessment (BELKA), an inaugural machine learning Kaggle competition aimed at addressing a critical challenge in drug discovery: predicting the likelihood of chemical materials binding to pharmaceutically relevant targets. The BELKA dataset, comprising 133 million physically measured activities for three key protein targets, exceeded the world's largest existing drug-target dataset, PubChem. Upon the competition's conclusion in July 2024, Leash planned to publicly release the full 3.6 billion physically measured interactions dataset, providing an invaluable resource for future drug discovery efforts worldwide.