IPECnet: ML model for predicting the area of water solubility of interpolyelectrolyte complexes

Abstract

Interpolyelectrolyte complexes (IPECs) are known for years as classic representatives of smart polymers. The solubility of IPECs in water–salt media is driven by numerous factors connected with polymer component parameters and media composition. This work is devoted to the development of the world's first machine learning-based model for predicting the area of existence of water-soluble IPECs for solving biomedical problems. A new approach is proposed that takes into account both the physico-chemical properties of polyelectrolytes and the chemical structures of their monomeric units. The developed approach is universal and can be used to predict the properties of multicomponent systems of a different chemical nature. The results of the work were applied to select the composition of water-soluble IPECs for treatment of surfaces in order to create bactericidal coatings. The dataset and model structures are publicly available on GitHub.