Leash Biosciences announced a successful funding round and launched an ML competition

The oversubscribed $9.3 million seed funding round was led by Springtide Ventures and backed by MetaPlanet, Top Harvest Capital, Mitsui Global Investment, MFV Partners, and Recursion co-founders Chris Gibson and Blake Borgeson. Leash Biosciences is working to develop "a foundational and generalizable machine learning model of medicinal chemistry that can accurately predict small molecule drug candidates for any protein in silico, and more broadly, interactions between any protein and any chemical." The company is compiling massive datasets of protein targets binding to chemicals to meet its goals. Thus far, its dataset comprises over 17 billion high-quality interaction measurements. The funding will enable Leash Biosciences to scale its data collection and computational capabilities to continue advancing its mission: the company plans to extend its dataset by screening over 500 protein targets against millions of proprietary chemicals designed by Leash's machine learning technology.

Jointly with its funding announcement, Leash Biosciences launched its inaugural machine learning Kaggle competition, aptly named the Big Encoded Library for Chemical Assessment (BELKA). In the competition, participants will be tasked with analyzing a dataset composed of a fraction of the company's total data: 133 million small molecule candidates were screened against three protein targets to produce a small molecule dataset larger than PubChem, the world’s largest existing drug-target dataset. Moreover, Leash plans to release the full dataset that comprises all the conditions and replicates aggregated for the contest dataset by the end of summer 2024. Registration for the BELKA competition is open until July 8, 2024. More information can be found here.