Neopoly AI applies causal machine learning to representation-based models to enhance extrapolation in complex data environments for drug discovery. The company offers Causal-Chemprop for small molecule property prediction and optimization, and Causal-ESM for enzyme property prediction using sequence-level representations. They also develop custom structural causal models integrated with client-specific AI models for targeted use cases.
Funding
Funding not disclosed
Founders
Product
Problem
Traditional material discovery methods are slow and inefficient, often relying on trial-and-error experimentation or computationally expensive simulations. Predicting the properties of novel molecules, especially those outside the scope of existing datasets, remains a significant challenge. This limits the ability to rapidly identify and design materials with desired characteristics for various applications.
Solution
Neopoly AI offers a platform that leverages causal inference and machine learning to accelerate the discovery of new materials. By incorporating causality into predictive models, Neopoly enhances their ability to extrapolate and accurately predict the properties of out-of-distribution molecules. This approach enables researchers to identify promising material candidates more efficiently and reliably than traditional methods. The platform facilitates the design of molecules with optimal properties, reducing the time and resources required for material discovery.
Target Audience
The primary target audience includes researchers and developers in the fields of drug discovery and materials science who are seeking to accelerate the identification and design of novel molecules.
Features
- Causal inference-based machine learning models for improved property prediction
- Enhanced extrapolation capabilities for out-of-distribution molecules
- Identification of optimal molecules based on desired properties