Our Generative AI-Driven Drug Discovery Platform accelerates drug discovery process, reduces costs, and enhances the efficiency of searching for the novel drug-like molecules through the following features:
Generation of New Ligands for a Given Protein
Identification of ligand-protein binding sites, molecular docking, de novo structure generation, scaffold growing, and intermolecular interactions analysis.
Annotation of Generated Compounds
– Predicting binding affinity;
– Evaluating ADMET properties.
NBG (Nature Based Generator) Approach for Designing Novel Ligands for Protein Targets
1
3D Structure of Binding Site
Site Radar/Site Selection
2
Grid Construction
Site Map
3
Ligand Structure Generation
Scaffold Generation/Scaffold Placement
4
Structures of Potential Ligands
Periphery Generation
Key Technical Advantages
A precise algorithm for ligand-protein binding sites identification
SiteRadar showed higher accuracy compared to Fpocket and PUResNet algorithms.
Accurate classification of ligand-protein binding sites
A high ability to predict the binding mode
A high correlation with experimental ligand-protein affinity data
A strong ability to distinguish between positions of different atom types
A precise algorithm for ligand-protein binding sites identification
SiteRadar showed higher accuracy compared to Fpocket and PUResNet algorithms.
Accurate classification of ligand-protein binding sites
A high ability to predict the binding mode
A high correlation with experimental ligand-protein affinity data
A strong ability to distinguish between positions of different atom types
Our platform workflow
Description of stages
Site Radar: prediction of active site from protein structure
Site Selection: bonding site selection based on ligands
Site Map: аrrangement probe
Scaffold Generation: generate structure of scaffold based on probes selected
Scaffold Placement: positioning a given 2D scaffold
Periphery Generation: generation of substituents around selected scaffold
