The study of quantitative structure-activity and structure-property relationships (QSAR/QSPR) and the computer molecular design of organic compounds with definite properties are among the most important problems of modern chemistry.

This research project is aimed at the development of novel approaches for the construction of models relating to the chemical structure and the properties of organic compounds, the development of novel descriptors of chemical structure for their application in such models, and the creation of novel methods for the solution of the inverse problem of the reconstruction of structures with the desirable range of property from the values of descriptors on the basis of the structure-property model equations.

Computer programs and approaches will be developed with the aim of studying the structure-property relationships on the basis of the artificial neural networks, including those with the original architecture, which will allow the direct correlation of the properties of organic compounds with their structure without the calculation of the descriptors.

The applicability of the methods suggested to the design of structures with definite properties will be illustrated for a variety of properties and for several classes of organic compounds.