Antonio Viayna and F. Javier Luque have recently published a collaborative work entitled "Prediction of Toluene/Water Partition Coefficient in the SAMPL9 Blind Challenge: Assessment of Machine Learning and IEF-PCM/MST Continuum Solvation Models" in the journal PCCP (Physical Chemistry Chemical Physics).
In this study, conducted within the context of the SAMPL9 challenge, they focused on dealing with a less commonly studied partition coefficient, the toluene/water one. This coefficient provides valuable insights into the propensity of molecules to form intramolecular hydrogen bonds and their chameleonic properties, which greatly influence solubility and permeability. Thanks to that they developed and compared two Machine Learning models: Multiple Linear Regression (MLR) and Random Forest Regression (RFR). Additionally, they performed a parameterization of the IEF-PCM/MST method to accurately predict solvation free energies of over 160 compounds in both toluene and benzene.
This project is the result of a big and extensive effort, involving collaborative work among different institutions, namely the University of Costa Rica (with Profs. William Zamora and Silvana Pinheiro), University of Barcelona (Profs. Carles Curutchet and Clara Ràfols), and Pion Inc (Dr. Rebeca Ruiz)
If you want more details about the article, check the link: https://pubs.rsc.org/en/content/articlelanding/2023/cp/d3cp01428ba