Nonlinear mapping techniques for predicting the pharmacological properties of chemical compounds

The state of the art of studies in the development and application of a special group of data analysis methods, namely, mapping algorithms, which allow decreasing the dimensionality of multidimensional spaces of the properties of analyzed objects (series of chemical compounds) to their two- or three-dimensional projections (images) with retention of the initial space topology is considered. It is shown how one can use such maps to visually analyze complex relationships between the properties and structural features of compounds, which is used for predicting the pharmacologically significant properties of chemical compounds. Bibliography - 86 references.