Overview
We help move from target to candidate, ranking and filtering compounds before they reach the bench.
Salient features
- Docking and rescoring of small molecules and peptides
- ADMET and property prediction
- Free energy and generative design options
Analyses we offer
Drug design
Combines structure and ligand information with machine learning and generative models to propose and prioritise compounds.
Methods: deepchem, rdkit, REINVENT4.
Docking of Peptides
Predicts how peptides bind to their targets and ranks candidate binders.
Methods: haddock3, boltz.
Docking of Small Molecules
Docks and rescores ligand libraries against a binding site for virtual screening.
Methods: AutoDock-Vina, gnina, DiffDock.
Protein-ligand Interactions
Profiles hydrogen bonds, hydrophobic contacts and other interactions in complexes and over simulations.
Methods: plip, ProLIF.
ADMET Analysis
Predicts absorption, distribution, metabolism, excretion and toxicity properties of compounds.
Methods: admet_ai, TDC.
QSAR / molecular property prediction
Builds models that predict activity or properties from chemical structure.
Methods: chemprop, deepchem.
Retrosynthesis planning
Proposes synthetic routes to a target molecule from available building blocks.
Methods: aizynthfinder.
Molecular descriptors & fingerprints
Generates descriptors and fingerprints for similarity search and modelling.
Methods: datamol, mordred.
Free energy perturbation (binding affinity)
Estimates relative binding affinities of related ligands with alchemical free energy calculations.
Methods: openfe, perses.
What you receive
- A written report with methods, parameters and interpretation
- Processed data tables and publication-quality figures
- Scripts and settings so every result can be reproduced
Not sure which analysis fits your data? Contact us and we will advise.