Technologies for drug discovery
Ligand based
Clustering
MPO
Multi parameter lead optimization (MPO)
- ML/DL-based models predict DMPK parameters and help prioritize syntheses
- Multi-parameter optimization driven effort to realize the evolving TPP in the series
Virtual Screening
Virtual Screening
Large collection of commercially available compounds for screening
- Structure based virtual screening (SBVS)
- Ligand based virtual screening (LBVS)
Large collection of commercially available compounds for screening
- Structure based virtual screening (SBVS)
- Ligand based virtual screening (LBVS)
Ultra Large Library Generation
Ultra Large Library Generation
Generating a large virtual library of compounds using diverse fragments
Generative Chemistry
Generative Chemistry
- Structure-aware molecule generation
- De novo design
- Superstructure generation
- Scaffold decoration
- Motif extension
- Linker generation
- Scaffold morphing
- Pocket-conditioned diffusion models
- Property-guided molecule optimization
- LLM-driven chemical ideation
- Evaluation and filtering
DFT
Density Functional theory (DFT)
- Lowest energy conformation of ligands
- Understand electronic properties and reactivity of a compound through Frontier Molecular Orbitals (FMO) approach
- Novel isostere identification through electronic properties
SAR
Structure Activity Relationship and Molecular Matched Pairs
Structure activity relationship (SAR) showcased through a variation of R group
MMP
Structure Activity Relationship and Molecular Matched Pairs
Structure activity relationship (SAR) showcased through a variation of R group
Conformational Search
Conformational search
3D conformers generation
Energy profile of the generated conformers
R-group Analyses
R-group Analyses
Structure activity relationship (SAR) showcased through a variation of R group
DEL/ASMS Screen Planning
DEL/ASMS Screen Planning
- DEL/ASMS screens expand druggable target space by offering chemical starting points
- No, we don’t do DEL/ASMS screens but we can plan it for you and connect you appropriately
- This is the workflow we adopt:
Pharmacophore
Pharmacophore
- Consensus and structure based pharmacophores
- 3D QSAR
- Screening against large library of compounds for a target protein (Ligand based virtual screening – 300M+)
Structure based
Pocket Druggability
Pocket Druggability
- Pocket detection and segmentation
- Druggability scoring and ranking
- Binding site classification
- Ligandability prediction
- Dynamic pocket analysis
- Cross-target pocket comparison
Clustering
InteractogramTM
InteractogramTM
InteractogramTM : Where the ligand meets the interaction hotspots – IFP
HydrogramTM
HydrogramTM
Identifying structural water sites in the binding pocket
Molecular Docking
Molecular docking
- Ensemble docking grid to account for the flexibility of the binding site residues
- Covalent, peptide and protein-protein, RNA docking
Covalent Docking
Covalent Docking
- Targeting suitable residues with covalent warheads
- Trap a range of reactive residues – Cys, Tyr, Lys, Ser, His
- Warhead electrophilicity assessment and modulation
Protein-Protein Docking
Protein-Protein Docking
Identify probable binary complexes using energetics, clustering and ML
Virtual Screening
Virtual Screening
Large collection of commercially available compounds for screening
- Structure based virtual screening (SBVS)
- Ligand based virtual screening (LBVS)
Large collection of commercially available compounds for screening
- Structure based virtual screening (SBVS)
- Ligand based virtual screening (LBVS)
Molecular Dynamics
Molecular Dynamics Simulations
Physics based simulations to identify preferred protein states to yield novel designs
RNA Docking
RNA Docking
- RNA structure preparation
- Binding site identification
- Ligand docking
- Protein–RNA docking
- Scoring and ranking
- Conformational flexibility
- RNA-quadruplex modeling
PROTAC Modeling
PROTAC Modeling
- Ternary complex generation for novel PROTACs
- Design and prospective modeling of PROTACs
Receptor Flexible Docking
Receptor Flexible Docking
- Ensemble docking grid to account for the flexibility of the binding site residues
- Covalent, peptide and protein-protein, RNA docking
Free Energy Estimation
Work in progress…
Peptide Design
Peptide design
- Proteins with flat and shallow binding pockets
- Targeted through linear/cyclic peptide design
- Improving the druggable target space
- Deep Learning based peptide design
- Macrocyclic peptides
- Interaction hotspots using XAI
FBDD
Fragment based drug discovery (FBDD)
- Identifying interaction hotspots in the binding site
- New ligand designs through hybridization of fragments
- In-situ ligand editing
AI/ML
Deep Learning
Deep Learning
Structure prediction
- Sequence → structure prediction
- Multimer / homomer / heteromer generation
- Peptide generation
- Complexes with ligands / nucleic acids
- Conformational Ensemble Prediction
- Mutagenesis experiment
Patent digitization
- Structure recognition and extraction
- Data integration and cross referencing
- Text and data extraction
- Dataset from patents
Molecular designs
- Structure-aware scaffold decoration
- Multi-anchor decoration
- R-group generation with variable size
- Pocket-conditioned generation
- End-to-end diffusion-based generation
Property prediction
- ADMET and drug-likeness prediction
- Tautomer / protonation state enumeration
- Micro-pKₐ Prediction
- Binding affinity estimation
- Building DL models
LLM
Clustering
Generative Chemistry
Generative Chemistry
- Structure-aware molecule generation
- De novo design
- Superstructure generation
- Scaffold decoration
- Motif extension
- Linker generation
- Scaffold morphing
- Pocket-conditioned diffusion models
- Property-guided molecule optimization
- LLM-driven chemical ideation
- Evaluation and filtering
Knowledge Graphs
Knowledge Graphs
Refining vast information into decisive knowledge graphs