Solubility Prediction from Molecular Properties and Analytical Data Using an In-phase Deep Neural Network (Ip-DNN)

Abstract

Materials informatics is an emerging field that allows us to predict the properties of materials and has been applied in various research and development fields, such as materials science. In particular, solubility factors such as the Hansen and Hildebrand solubility parameters (HSPs and SP, respectively) and Log P are important values for understanding the physical properties of various substances. In this study, we succeeded at establishing a solubility prediction tool using a unique machine learning method called the in-phase deep neural network (ip-DNN), which starts exclusively from the analytical input data (e.g., NMR information, refractive index, and density) to predict solubility by predicting intermediate elements, such as molecular components and molecular descriptors, in the multiple-step method. For improving the level of accuracy of the prediction, intermediate regression models were employed when performing in-phase machine learning. In addition, we developed a website dedicated to the established solubility prediction method, which is freely available at “”.

Materials informatics approach using domain modelling for exploring structure–property relationships of polymers

“…We have created a Bayesian optimization-based 50 spectral simulation tool that automates the T 2 relaxation time domain fitting, frequency fitting, and 3D domain modeling of our domain component separation method. The details of the Python program including T 2 relaxation time information, frequency information, and 3D domain modeling of the present domain component separation method can be obtained at https://github.com/riken-emar/matrigica .…”

“…Generative topographic mapping regression (GTMR) has also been applied to the analysis of CP-MAS spectra to predict 13 C NMR spectrum of the material in its solid-state based on its thermophysical properties 34 . Machine learning methods have applied for various material studies such as cloud-point engineering of polymers 47 , prediction of drug-polymer amorphous solid dispersion miscibility and stability 48 , atomic/inter-atomic properties prediction 49 , solubility prediction 50 , descriptor selection for investigating physical properties of biopolymers in hairs 51 , classification of the membrane materials 52 , prediction of crystallization tendency 53 , prediction of density, glass transition temperature, melting temperature, and dielectric constants of polymer 9 , macromolecular modeling 54 .…”

Materials Informatics Approach Using Domain Modelling For Exploring Structure-Property Relationships Of Polymers

“…We have created a Bayesian optimization-based 50 S2. Red Mobile, Magenta Intermediate(Mobile), Violet Intermediate(Rigid), Blue Rigid.…”

The exposome paradigm to predict environmental health in terms of systemic homeostasis and resource balance based on NMR data science

“…We have successfully developed several ML-based analytical approaches, namely, a prediction method for metabolic mixture signals, 225,226 DNN-mean decrease accuracy, 32 ensemble DNN, 33 variable selection for regional feature extraction, 34 and methods for evaluation of surface water, 14 impact estimation of food intake on mice, 227 evaluation of human daily dietary intake 228 and relaxometric learning 229 in metabolomics studies. DNN has also been used to reconstruct non-uniformly sampled NMR spectra 230,231 and for solubility prediction, 232 while CNN has been used to remove noise from NMR spectra 233 and to reconstruct NMR spectra from non-uniformly sampled data. 234 DN-Unet combines structures of encoders–decoders, and CNN can be used to suppress noise in liquid-state NMR spectra to enhance SNR.…”