Abstract
A random forest model obtained from calculated physicochemical properties of solvents and observed crystallised structures of olanzapine has for the first time enabled the prediction of different types of 3-dimensional crystal packings of olanzapine solvates. A novel olanzapine solvate was obtained by targeted crystallization from the solvent identified by the random forest classification model. The model identified van der Waals volume, number of covalent bonds and polarisability of the solvent molecules as key contributors to the 3-D crystal packing type of the solvate.
Original language | English |
---|---|
Pages (from-to) | 3947-3950 |
Number of pages | 4 |
Journal | CrystEngComm |
Volume | 20 |
Issue number | 28 |
DOIs | |
Publication status | Published - 24 Apr 2018 |
Keywords
- random forest model
- crystal packing
- olanzapine solvates
- physicochemical properties
- Machine Learning