Glassomics: An omics approach toward understanding glasses through modeling, simulations, and artificial intelligence

Glass science, like other materials domains, has been advancing at a rapid pace during the last few decades thanks to sophisticated experimental techniques, simulation methods, and computing capabilities. Specifically, the availability of a plethora of information about glass compositions and properties, synthesis and experimental protocols, and large-scale and multiscale modeling has enabled a unified approach to study glasses through a bottom-up approach. Here, taking inspiration from the “omics” approach of biological science, we propose “glassomics” to study glasses in a holistic fashion. We discuss how the advances in artificial intelligence, experiments, and simulations allow a high-throughput screening of glasses while scanning the entire periodic table-based compositions. We discuss how glassomics can provide a comprehensive understanding of complex interrelationships of composition, structure, process, and properties of glasses through simulations, machine learning, and natural language processing by reviewing the latest trends in the field. Finally, we also discuss some of the outstanding challenges in the field and some of the potential approaches toward addressing them. Graphical abstract: [Figure not available: see fulltext.]