Artificial Intelligence & Machine Learning (AI/ML)
Accelerating discovery with predictive modelling services that incorporate the latest advances in AI and Machine learning, guided by Domainex expertise.
- Build custom AI/ML models tailored to your project
- Predict activity, selectivity, and absorption, distribution, metabolism, excretion and toxicity (ADMET) properties
- Identify non-obvious SAR
- Support data-driven decision making at every stage of discovery
Ligand-Based Drug Design
When structural data is limited, ligand-based drug design (LBDD) allows you to build predictive models and design promising scaffolds.
- Build predictive models from your ligand datasets
- Identify and optimise promising scaffolds
- Expand SAR exploration with focused compound sets
- Pharmacophore/Shape/Electrostatic Modelling
- Scaffold Hopping
- Leverage proprietary and commercial chemical databases for rapid design cycles
Structure-Based Drug Design
Exploit protein structures for precision design using Domainex’s structure-based drug design (SBDD) services.
- Molecular docking, pharmacophore modelling, and dynamics simulations
- Torsional analysis to identify strain in bound ligand conformations
- Binding site detection and druggability assessments – in combination with Molecular Dynamics (MD) enables the identification of allosteric/cryptic binding sites
- AI-enabled protein–ligand co-folding and traditional homology modelling when crystal structures are not available
- Cross-target structural comparisons to design selective ligands
- Water mapping to identify key hydration sites driving affinity and selectivity
- Accurate physics-based assessment of binding affinities via Free Energy Perturbation (FEP) and Fragment Molecular Orbital (FMO) Analysis
- AI-guided blind docking to discover novel binding pockets without prior assumptions
- Machine Learning (ML) approaches to identify potentially reactive cysteines
- Applied to diverse targets including:
ADMET Modelling & Property Prediction
Guiding compounds into the right property space, through an in depth understanding of Absorption, Distribution, Metabolism, Excretion and Toxicity.
- Prediction of efflux, Kp,uu, permeability, pKa and logD
- AI/ML enabled modelling of rates and sites of metabolism (CYP450 isoforms and hAOX)
- hERG, AMES and photoxicity modelling
- Early flagging of undesirable features
- Integration of computational predictions with experimental measurements