SuperCowPowers / sageworks

SageWorks: An easy to use Python API for creating and deploying AWS SageMaker Models
https://www.supercowpowers.com
MIT License
37 stars 1 forks source link

RDKIT Based Modeling and Documentation #168

Open brifordwylie opened 1 year ago

brifordwylie commented 1 year ago

Let's do an RDKIT based example that shows the process of using SageWorks for rapid prototyping.

References:

brifordwylie commented 6 months ago

ADME Properties: Absorption, Distribution, Metabolism, and Excretion. Knowing these can help optimize drug delivery and efficacy.

Toxicity: You'll definitely want to predict if a compound could be toxic to humans or other organisms.

Binding Affinity: Knowing how well a molecule will bind to a target protein can be key for drug design.

Stability: How long does the molecule last under various conditions? This impacts shelf life and effectiveness.

LogP: The partition coefficient between water and octanol. It gives you an idea about the drug's solubility in fat versus water, which affects how it'll be distributed in the body.

pKa: This property helps you understand how a molecule will behave at different pH levels, affecting its solubility and stability.

Molecular Weight: It often correlates with other important properties like permeability and solubility.

Tautomeric Forms: Some molecules can exist in multiple tautomeric forms, and these can have different properties.

Bioavailability: How much of the drug will actually make it into the bloodstream when administered in various forms (oral, IV, etc.).

Drug-Drug Interactions: If your molecule is likely to interact with other commonly-used drugs, that’s key info.

Resistance Profile: For antimicrobial drugs, how quickly might pathogens develop resistance?