This is the solubility prediction tool for calculating the solubility of small organic molecules in water and organic solvents, developed by the iPRD at University of Leeds.

Please note: this app is only recommended for use with the most up-to-date versions of Firefox, Chrome, Opera and Microsoft Edge.

Details of the protocol used can be found in this publication:

Samuel Boobier, David R. J. Hose, A. John Blacker, Bao N. Nguyen, "Machine Learning with Physicochemical Relationships: Solubility Prediction in Organic Solvents and Water", Nature Communications 2020, 11, 5753 (doi:10.1038/s41467-020-19594-z).

If you are interested in an improved version of this prediction tool, with integrated work-flow, in-house training data and batch predictions, please contact Dr Bao Nguyen directly.

How to Use this Tool

This cheminformatics tool calculates descriptors based on 3D structures optimised using DFT. In Step 1 the user must generate input files for a Quantum Mechanics software, Gaussian 09 from a compound of their choice. After the user has run these files on their own computer/HPC, Step 2 is to upload the output files to our server, where solubility is calculated with machine learning models.


To use this prediction tool, you will need:
1. Access to Gaussian 09.
2. Melting point of the target compound (optional, recommended for non-aqueous predictions).

Step 1: Create Input Files for Gaussian 09

Create Input Files

Step 2: Use Gaussian 09 Output Files to Predict Solubility

Predict Solubility

Disclaimer: The authors accept no liability for the topicality, correctness, completeness or quality of the provided predict or how the predictions are subsequently used. The authors expressly reserve the right to change, supplement or delete parts of the pages or the entire tool without separate announcement or to discontinue the tool temporarily or permanently.

Visitor counter: 1432