We present a new script developed by the staff of our center – OxidationStatePredictor, and a website based on it with a service for determining the oxidation state of metal atoms according to the CIF file - crystalpredictor.com.The service allows you to determine the degree of oxidation of metal atoms associated with oxygen in the structure of ionic and coordination compounds.
The service development team included Mikhail Smolkov, Igor Karpukhin and Dr. Alexander Shevchenko.
Using machine learning methods, our specialists for the first time developed a model that, based on geometric and topological descriptors, calculates the degree of oxidation of metal atoms in an oxygen environment. The model was trained and validated on the basis of information about 90 thousand metal atoms contained in 35 thousand crystal structures from the ICSD and CSD databases.