Computation
Physics-based simulation
- Enhanced MD Simulation
- Hot-spot Mapping
- HT Virtual Screening
- Free Energy Calculation
Artificial Intelligence
- Conformation Prediction
- Pocket Prediction
- Hit finding and Ligand Design
- DMPK Property Prediction
EnsemGen™
- Metastable states generation
- De Novo molecule design
Kinetic Ensemble®
EnsemNet™
- Cryptic pocket prediction
- Fine-tuned predictive apps
Experimentation
Biophysical Technologies
- HDX-MS
- Cryo-Electron Microscopy
- X-Ray Crystallography
- NMR dynamics (CEST+CPMG)
Chemical Biology
- Experimental Screen
Developing First-In-Class or Best-In-Class Medicines Faster, Better and Cheaper
Our Kinetic Ensemble® platform effectively integrates novel computational and experimental technologies to discover cryptic pockets on macromolecules and hit compounds for precision medicine approaches to drug discovery.
EnsemNET™
- A state-of-the-art deep learning architecture for predicting protein cryptic pockets, pocket similarity, protein-ligand binding and enabling SAR transfer learning.
- Pre-trained on large protein and ligand databases, EnsemNET™ continuously improves its accuracy by fine-tuning with newly published and proprietary datasets.
- Synergistic with equilibrium-state predictions by physics-based computational approaches.
EnsemGEN™
- An advanced generative neural network designed to predict protein kinetic metastable states, transient pockets and generate de novo ligands with desired properties.
- EnsemGEN™ was trained on vast protein conformations and ligand libraries to capture protein dynamics and intricate protein-ligand interactions.
- By harnessing the power of generative AI, EnsemGEN™ reveals uncharted protein functional states and uncovers new pockets for unique drug design strategies.