Introduction

This model contains a machine-learning algorithm that was trained to predict the glass transition temperature T_g of molecular organic substances. The model provides two different modes of operation depending upon the availability of input parameters describing the molecular structure.
The SMILES Mode requires information on the molecular structure in the format of the Simplified Molecular-Input Line-Entry System (SMILES), which is a standardized method to describe the chemical structure of a molecule in a single string. To generate a SMILES string from a chemical structure, several independent tools are available such as the SMILES generator.
The other version, the Functional Group Mode, can be used without knowledge of a SMILES string, which, however, goes along with a reduced accuracy in the predicted T_g. For further information on the physical background of the model, we refer to the following paper: https://doi.org/10.1021/acsomega.2c08146
Please choose a mode from the navigation bar above. We recommend the SMILES Mode for higher accuracy.