ChemBio-SAR
- Generative Chemistry (GC) – Generation of potent, selective, and drug-like compounds with graph-based transformers in a Reinforcement Learning framework
- In vivo and in vitro ADME-Tox models to reinforce the transformers in GC
- Empowering the discovery by GC and ADME-Tox models-driven proprietary database
- Unlocking new possibilities with advanced biological models for selectivity and polypharmacology
- Combining the AI models with computational chemistry to create a new era in drug discovery
Drug-X
- Recurrent Geometric Network (RGN) for end-to-end modeling of orphan proteins
- Unlocking the proteins by identifying the druggable pockets with mixed solvent sampling, and evolutionary methods
- Fragment-based scanning and geometry-based fingerprint generation for high-speed virtual screening of large databases
- Discovery of high-affinity molecules for difficult targets utilizing GC & Molecular Dynamic Simulations (MDS)
- Docking, MDS, water thermodynamics, and FEP for prioritizing the desired compounds
MedChem
- Improving the desired properties and synthetic feasibility of the selected molecules
- Synthetic route estimation using big data
- Reaction mechanism estimation for solving complex bottlenecks with better yields
- Bioisostere replacement for improving the druggable properties
- Impurity predictions for a better understanding of the synthetic protocols
Platform Highlights
- Accurate protein structure prediction for complicated and orphan targets
- Identifying the cryptic pocket and druggability analysis
- AI-generated large proprietary drug-like database
- Flexible framework applicable to any target
- Automated multi-parameter optimization with integrated AI and computational techniques
Platform Advantages
- Integrated AI and CADD platform to generate drug candidates with less turnaround time
- Generates quick hypothesis with validation
- Rapid virtual screening AI protocols
- Coordination of in-house AI, MedChem, and Synthetic teams to convert the virtual compounds into reality reduces cost and time
- Accurate prediction of efficacy, selectivity, and safety
Case studies
- AxDrug® applied for 30 different targets with a 100% success rate in discovering drug candidates
- Identified first-in-class molecules for the most intricate targets
- Successfully innovated best-in-class molecules with polypharmacology
- Modeled 3D structures and generated drug candidates forhighly disordered proteins
- POC established for each program by synthesizing ~ 200 molecules