Leash Bio is developing a proprietary dataset of protein-small molecule interactions by screening millions of compounds against thousands of proteins, generating billions of data points for machine learning applications in drug discovery. This approach addresses the lack of comprehensive data in medicinal chemistry, enabling faster identification of potential drug candidates for oncology and other diseases.
Funding
$9.4M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
SVFounders
Product
Problem
Drug discovery is hindered by a lack of comprehensive data on protein-small molecule interactions, making it difficult to identify promising drug candidates efficiently. Existing medicinal chemistry data is often incomplete, limiting the effectiveness of machine learning applications in predicting drug efficacy and safety.
Solution
Leash Bio is creating a proprietary dataset of protein-small molecule interactions to accelerate drug discovery. The company screens millions of compounds against thousands of proteins, generating billions of data points that are ideal for training advanced machine learning models. This rich dataset enables the rapid design of novel chemical matter with a higher probability of having the desired biological activities. Leash Bio's approach involves a dynamic, cyclical engine that continuously harnesses data, iterates machine learning models, and refines its approach to identify hits. The company is applying this technology to develop new medicines internally, starting with oncology targets, and is also partnering with biopharma companies to explore new molecule opportunities.
Target Audience
Leash Bio's primary customers are biopharma companies seeking to accelerate their drug discovery programs, as well as researchers and drug developers focused on oncology and other diseases.
Features
- A dataset of 30 billion protein-molecule interactions measured.
- 6.7 million ML-designed molecules made and tested.
- Scalable software for rapid design of new molecules.
- Cyclical engine that continuously harnesses data and iterates machine learning models.
- Platform to measure 500 protein targets against 20M molecules each in the next year.