Neural networks as a method for elucidating structure–property relationships for organic compounds

Abstract

The published data devoted to the use of the neural The published data devoted to the use of the neural network approach in the simulation of structure ± property rela-network approach in the simulation of structure ± property relationships for organic compounds are reviewed. The basic princi-tionships for organic compounds are reviewed. The basic principles of the neural network simulation are discussed along with the ples of the neural network simulation are discussed along with the characteristic features of the neural network approach typical of characteristic features of the neural network approach typical of the representation and classification of structural chemical data. the representation and classification of structural chemical data. Brief information on neural network models of spectral character-Brief information on neural network models of spectral characteristics, reactivities, physicochemical properties and biological istics, reactivities, physicochemical properties and biological activities of organic compounds is presented. The bibliography activities of organic compounds is presented. The bibliography includes 159 references includes 159 references. . Neural networks consist, as a rule, of a set of relatively simple computing elements termed neurons; the latter are arranged in several layers and are linked to one another by numerous connections. The layers are usually subdivided into three groups, viz., input, hidden and output (Fig. 1). Each connection between two neurons is defined by a real number termed connection

A renaissance of neural networks in drug discovery

“…17,18 When ANNs are applied to drug discovery, the modeled properties are often physicochemical and ADMET properties of organic compounds, toxicity endpoints, binding constants or IC 50 values with respect to various macromolecular biological targets, types and profiles of biological activity, etc. (e.g., see comprehensive tables in the review article 13 ).…”

“…11 For the last 25 years this approach to modeling structure-activity relationships has matured into a well-established scientific field with numerous theoretical approaches and successful practical applications (see review articles [12][13][14][15][16] ). The field now encompasses the use of ANNs for predicting not only different types of biological activity but also physicochemical, Absorption Distribution Metabolism Excretion and Toxicity (ADMET), biodegradability and spectroscopic properties, and reactivity.…”

Fragment Descriptors in SAR/QSAR/QSPR Studies, Molecular Similarity Analysis and in Virtual Screening

“…Among nonlinear regression methods used in conjunction with fragment descriptors, the Back-Propagation Neural Networks (BPNN) [286][287][288][289] occupy a special place. It has been proved 7,8 that any molecular graph invariant can be approximated by an output of a BPNN using fragment descriptors as an input.…”

Chemometrics: achievements and prospects

“…2 QSAR studies related to chemometrics headed by Academician Zefirov are underway. 50 The metrological aspects and control of the quality of chemical analysis are investigated in Dvorkin's works. 51 A research group headed by Vlasov 52 at the St Petersburg State University is working on sensor systems known as thè electronic tongue', and analogous systems called 'electronic nose' are developed at the Voronezh Technological Academy.…”