Solubility of organic compounds in octanol: Improved predictions based on the geometrical fragment approach

- Two very simple models to predict octanol solubility are reported.
- In the lack of experimental data, the first one performs as well as current state-of-the-art models.
- Using melting point data, the second one yields some improvement through simple corrections.
- Good performance is achieved through a physically-motivated fragmentation of the molecules.
- Some limitations of popular applicability domain definitions are demonstrated.

Two new models are introduced to predict the solubility of chemicals in octanol (Soct), taking advantage of the extensive character of log(Soct) through a decomposition of molecules into so-called geometrical fragments (GF). They are extensively validated and their compliance with regulatory requirements is demonstrated. The first model requires just a molecular formula as input. Despite an extreme simplicity, it performs as well as an advanced random forest model involving 86 descriptors, with a root mean square error (RMSE) of 0.64 log units for an external test set of 100 molecules. For the second one, which requires the melting point Tm as input, introducing GF descriptors reduces the RMSE from about 0.7 to <0.5 log units, a performance that could previously be obtained only through the use of Abraham descriptors. A script is provided for easy application of the models, taking into account the limits of their applicability domains.