LogP Prediction
The logarithm of the octanol–water partition coefficient, logP, is a crucial quantity for small-molecule therapeutics. LogP measures a compound's lipophilicity, or "fattiness," which correlates with its ability to dissolve, be absorbed through the intestinal wall, and permeate cell membranes. A well-balanced logP is crucial for an effective drug: greasy molecules with a high logP are often poorly soluble and can bind or accumulate in unwanted places, while polar molecules with a low logP are often poorly bioavailable and cell-permeable.
Rowan's logP workflow offers three different methods for calculating or predicting logP, ranging from fast cheminformatic models to a first-principles physics-based approach:
- RDKit's Crippen model is an atom-contribution-based method that's been a mainstay of industry cheminformatics for decades. It's essentially instant and a great default choice.
- Our Chemprop D-MPNN model is a machine-learning model trained on the recently released SangsterLogP dataset of 23,000 unique experimental logP values.
- Our physics-based COSMO-RS method calculates logP from first principles rather than pattern-matching against training data. It's much slower, but a useful option for complex or out-of-distribution molecules like unusual zwitterions, exotic salts, macrocycles, or heavily fluorinated compounds that are poorly represented in any experimental dataset.
How to run logP predictions on Rowan
Choose a molecule, choose your preferred method—or run all three side by side to compare—and submit. Results appear in Rowan's browser-based interface within seconds for the cheminformatic and ML methods, or within a few minutes for the physics-based COSMO-RS method.
Ready to try it out? Create a free Rowan account and predict logP for your molecules today.
