Causal Machine Learning for
Drug DiscoveryIntroducing causality to representation-based models for improved extrapolation in challenging data regimes.
Our Technology
⌬ Causal-Chemprop
SCM built on Chemprop representations and additional molecular descriptors for small molecule property prediction and optimization.
🧬 Causal-ESM
SCM built on ESM-2 sequence-level representations and additional physicochemical descriptors for enzyme property prediction.
🔬 Custom
We’ll build a structural causal model using your in-house AI model for your use case.
Publications
Causal-Chemprop: Causal Molecular Machine Learning for Property Prediction and Molecular Optimization
Christian Natajaya (1), Lucas Attia (2), Jackson Burns (2), Patrick Doyle (2)
(1) Neopoly Ltd, London, United Kingdom
(2) Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA