openOCHEM consensus model wins Kaggle First EUOS/SLAS Joint Compound Solubility Challenge

ChemRxiv

Abstract

The EUOS/SLAS challenge has its goal to develop reliable algorithms to predict solubility of small molecules experimentally measured aqueous solubility of 100k compounds. In total, hundred teams took part in the challenge to predict low, medium and highly soluble compounds as measured by nephelometry assay. This article describes the winning Top I rank model which was developed using the publicly available Online CHEmical database and Modeling environment (OCHEM) and which is publicly available at https://ochem.eu/article/​​27. We describe in detail the assumptions and steps used to select methods, descriptors and strategy which contributed to the winning solution.

Authors 4

1. Peter B. R. Hartog

   Helmholtz Munich

   Affiliation as printed

   Helmholtz Zentrum München

2. Martin Šícho Aachen

   Leiden University · University of Chemistry and Technology, Prague

   Affiliation as printed

   Leiden University

   University of Chemistry and Technology

3. Guillaume Godin

   DSM-Firmenich AG (Switzerland)

   Affiliation as printed

   DSM-Firmenich SA

4. Igor V. Tetko

   Helmholtz Munich

   Affiliation as printed

   Helmholtz Zentrum München