Prediction of Solid State Properties of Cocrystals Using Artificial Neural Network Modeling

Abstract

Using Artificial Neural Networks (ANNs), four distinct models have been developed for the prediction of solid-state properties of cocrystals: melting point, lattice energy, and crystal density. The models use three input parameters for the pure model compound (MC) and three for the pure coformer. In addition, as an input parameter the model uses the pK a difference between the MC and the coformer, and a 1:1 MC−conformer binding energy as calculated by a force field method. Notably, the models require no data for the actual cocrystals. In total, 61 CCs (two-component molecular cocrystals) were used to construct the models, and melting temperatures and crystal densities were extracted from the literature for four MCs: caffeine, theophylline, nicotinamide, and isonicotinamide. The data set includes 14 caffeine cocrystals, 9 theophylline cocrystals, 9 nicotinamide cocrystals, and 29 isonicotinamide cocrystals. Model-I is trained using known cocrystal melting temperatures, lattice energies, and crystal densities, to predict all three solid-state properties simultaneously. The average relative deviation for the training set is 2.49%, 6.21%, and 1.88% for the melting temperature, lattice energy, and crystal density, respectively, and correspondingly 6.26%, 4.58%, and 0.99% for the valdation set. Model-II, model-III, and model-IV were built using the same input neurons as in model-I, for separate prediction of each respective output solid−state property. For these models the average relative deviation for the training sets becomes 1.93% for the melting temperature model-II, 1.29% for the lattice energy model-III, and 1.03% for the crystal density model-IV, and correspondingly 2.23%, 2.40%, and 1.77% for the respective validation sets.

Cocrystal Prediction Using Machine Learning Models and Descriptors

“…In recent years, machine-learning (ML) has emerged as promising tool for data-driven predictions in pharmaceutical research, such as quantitative structure-activity/property relationships (QSAR/QSPR), drug-drug interactions, drug repurposing and pharmacogenomics [39]. In the area of pharmaceutical cocrystal research, Rama Krishna et al applied artificial neural network to predict three solid-state properties of cocrystals, including melting temperature, lattice energy, and crystal density [40]. Przybylek et al developed cocrystal screening models based on simple classification regression and Multivariate Adaptive Regression Splines (MARSplines) algorithm using molecular descriptors for phenolic acid coformers and dicarboxylic acid coformers, respectively [41].…”

Two models to estimate the density of organic cocrystals

“…Krishna et al 11 have developed a model for predicting the density of cocrystals using artificial neural network based on some descriptors, such as mass weight, binding energy, melting point, and p K a .…”

A Universal Deep Learning Framework based on Graph Neural Network for Virtual Co-Crystal Screening

“…Several works already utilized the ML methods to conduct meaningful attempts in the cocrystal field. For example, artificial neural network (ANN) was used to predict melting point, density and lattice energy of cocrystals 18 20 . They calculated 1444 descriptors for each co-former, and used the absolute value or square of the difference between the two co-former descriptors as the sample representation.…”