Variational AI has developed Enki, the first commercially available foundation model for small molecules, utilizing a generative AI framework that requires no prior data input. This platform enables biopharma partners to rapidly generate novel and selective lead structures for over 300 GPCR and kinase targets, significantly accelerating the drug discovery process.
Funding
$3.6M 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.
FFFounders
Product
Problem
Traditional drug discovery methods for small molecules are slow, expensive, and often require extensive experimental data, limiting the ability to rapidly identify novel lead structures, especially for challenging targets like GPCRs and kinases. Existing computational approaches may struggle with targets lacking sufficient data or require significant manual curation.
Solution
Variational AI's Enki platform addresses these challenges by providing a commercially available foundation model for small molecule drug discovery. Enki utilizes a generative AI framework that does not require prior data input, enabling the rapid generation of novel and selective lead structures. The platform allows biopharmaceutical partners to define their target product profile (TPP) and then leverages an ensemble of generative algorithms trained on decades of experimental data to create potential drug candidates. This approach significantly accelerates the drug discovery process, offering access to over 300 GPCR and kinase targets.
Target Audience
The primary target audience consists of biopharmaceutical companies and research organizations involved in small molecule drug discovery, particularly those focusing on GPCR and kinase targets.
Features
- Generative AI framework for de novo small molecule design without requiring target-specific training data
- Target Product Profile (TPP) definition interface for specifying on-target, off-target, and desired molecular properties
- Ensemble of generative algorithms trained on extensive experimental data
- Access to over 300 GPCR and kinase targets, with ongoing expansion
- Rapid generation of novel and selective lead structures in approximately one to two weeks