Neural Network Prediction Model of Cocrystal Melting Temperature Based on Molecular Descriptors and Graphs

Abstract

Among the physical properties characterizing cocrystals, melting temperature is one of the primary properties. Its prediction has been done by researchers, but in the known prediction models, there are some limitations. In order to adapt to the requirements of data development further and improve the quality of prediction, two prediction models of cocrystal melting temperature, the artificial neural network (ANN(I)) and the message passing neural network (MPNN), are proposed in this paper with molecular descriptors and molecular graphs as the inputs. As for ANN(I), molecular descriptors are applied to the prediction of cocrystal melting temperature for the first time, and the importance of features is evaluated by SHapley Additive explanation. For dataset 1 established in this paper, the prediction accuracy of ANN(I) is 95.75% on the training set and 95.67% on the test set. The feature evaluation of ANN(I) reveals that GATS2v, AMR, BCUTc-1h, MLFER_BH, BCUTp-1l, and ATSC1c play major roles in predicting the melting temperature. The ANN(I) model based on molecular descriptors could explain the relationship between the molecular structure and melting temperature to some extent, but its prediction accuracy is relatively poor. Thus, the message passing neural network (MPNN), a deep learning model based on molecular graphs, is proposed. For the test sets of two datasets used in this paper, the prediction accuracy of MPNN is 99.84 and 99.63%, respectively. Also, in the data pre-processing, the MPNN extracts the structural features of molecules through molecular graphs, which could save time and workload a lot. The excellent performance suggests that the MPNN model might be a more suitable tool for the prediction of cocrystal properties in the future.

Molecular Dynamics Simulations of a Hexanitrohexaazaisowurtzitane/4‐Amino‐3,7‐dinitro‐ [1,2,4]triazolo[5,1‐c] [1,2,4]triazine Cocrystal

“…The intermolecular forces (e. g., Van der Waals force, hydrogen bond, π–π accumulation) play an important role in the formation of the cocrystal [18] . Among them, the bond energy of hydrogen bond is much larger than other intermolecular forces, and is the most important force.…”

“…Construction of the crystal model [17] The intermolecular forces (e. g., Van der Waals force, hydrogen bond, π-π accumulation) play an important role in the formation of the cocrystal. [18] Among them, the bond energy of hydrogen bond is much larger than other intermolecular forces, and is the most important force. CL-20 contains polar NO 2 -groups and TTX contains NH 2 -groups, which could form NÀ O•••H type hydrogen bonds.…”

Machine Learning Prediction of Flavonoid Cocrystal Formation Combined with Experimental Validation

“…The reliability of this method has been proved in the previous work. 17 To help find the scientific basis for selecting molecular descriptors, the QSPR method was used to select molecular descriptors. 64 As analyzed in the data set, the formation of flavonoid cocrystal is closely related to hydrogen bonding.…”

“…The QSPR method mainly uses various statistical methods and molecular structure parameters to study the relationship between the structure of compounds and their various physical and chemical properties and has been widely used in the field of molecular property prediction. The reliability of this method has been proved in the previous work . To help find the scientific basis for selecting molecular descriptors, the QSPR method was used to select molecular descriptors …”

“…They now have a wide range of applications in multiple areas, particularly drug discovery and materials science. − The cocrystal prediction models established based on machine learning algorithms generally follow the knowledge-based method and the physics-based paradigm. Their main idea is to complete the model construction by combining various descriptors or molecular fingerprints through big data learning or extraction of features in cocrystal samples. − Among these models, traditional machine learning algorithms such as support vector machine (SVM), random forest (RF), and logistic regression (LR) algorithms provide useful guidelines for experimental design due to their more explicit scientific basis. However, compared with artificial neural networks (ANNs) and deep learning, they tend to show only moderate predictive accuracy .…”