Prediction of the solubility of molecules

This is a reproduction of the work by Delaney.

In this project linear regression is applied to predict the solubility of molecules. To achieve that we use the SMILES notation of the molecules in order to obtain some molecular descriptors of interest using rdkit.

Specifically, the 4 molecular descriptors employed to predict the solubility (LogS) in the linear regression model are:

• cLogP: Octanol-water partition coefficient.
• MolWt: Molecular weigth.
• RB: Number of rotatable bonds.
• AP: Aromatic proportion = number of aromatic atoms / number of heavy atoms.

After building the linear regression model, the pycaret module is employed to compare the performance of several machine learning models at once.

The code for this project can be found here.