Chemistry42 | Insilico Medicine

Harness the power of Generative AI and Physics-based methods

Chemistry42 is a comprehensive small molecules drug discovery platform. It combines both the flexibility of generative AI and accuracy of physics-based methods to create optimal molecules.

Applications

Generative Chemistry

• Create novel small molecules with optimized properties using generative AI: de novo design, hit optimization, scaffold hopping, and R-groups search.
• Predict and optimize physicochemical and ADMET molecular properties.

ADMET Profiling

• Predict kinome activity to find off-targets and build selectivity profiles.
• Accurately estimate the relative binding free energy to prioritize molecules with efficient physics-based methods.

A Diverse Toolbox for Drug Discovery

• Hit Identification: Fully de novo generative chemistry, Virtual Screening.
• Hit-to-Lead: R-group exploration, Scaffold hopping.
• Lead Optimization: ADMET optimization, Golden Cubes, Alchemistry.

Workflows

1. R-Group Exploration
2. Scaffold Hopping
3. Potency and Selectivity Optimization
4. Generative ADMET Optimization

ADMET Profiling:

• Predict and optimize the ADMET profile of your lead molecules. Use our ADMET models as stand-alone tools or with Generative Chemistry to guide the design towards better molecules.

Golden Cubes

Predict kinome activity and selectivity.

• Golden Cubes works with 2D and 3D structures, using models trained on carefully curated activity datasets.
• Upload your target protein and/or ligand.
• Specify a pharmacophore hypothesis.
• Select an anchor hypothesis to preserve 3D fragments.
• Choose a target compound profile. Use Chemistry42GPT for help.
• Define desirable ADMET profile.
• Track your experiment's progress.
• Review and filter the results, prioritize your lead compounds.
• Visualize building blocks.
• Visualize generated molecules in the binding site.