PETIMOT: An AI that Predicts Protein Dynamics, Incredibly Fast

This article covers three recent AI developments in drug discovery. PETIMOT uses an equivariant graph neural network to quickly predict protein dynamics from static 3D structures, offering an efficient tool for finding cryptic pockets and allosteric sites. Another study shows how an interpretable machine learning model can accurately predict inhibitor selectivity and provide chemical design ideas. Lastly, SiMGen achieves zero-shot molecular generation through a local similarity principle, challenging the reliance on big data.

AI Drug Discovery
Graph Neural Network
Interpretable AI
Molecular Generation
Protein Dynamics
Author

Jixing Liu, Ph.D.

Published

Friday, the 19th of September, 2025

Table of Contents

  1. PETIMOT is an AI framework based on a graph neural network. It can rapidly and accurately infer complex protein dynamics directly from sparse, static 3D structures, providing a transformative tool for conformational analysis in drug discovery.
  2. A study demonstrates that a simple, interpretable AI model capable of assessing its own prediction confidence can effectively predict the selectivity of hCA inhibitors and offer practical molecular design ideas.
  3. SiMGen uses a local similarity approach to show that generating complex molecules doesn’t require massive datasets, offering a more flexible and controllable new path for drug design.

1. PETIMOT: An AI that Predicts Protein Dynamics, Incredibly Fast

In drug discovery, protein dynamics are critical. Proteins aren’t static structures; they constantly twist and fold. Capturing the specific conformations that allow a

Back to top