Special Issue: Chemoinformatics

Abstract

Chemoinformatics techniques were originally developed for the construction and searching of large archives of chemical structures but they were soon applied to problems in drug discovery and are now playing an increasingly important role in many additional areas of chemistry. This Special Issue contains seven original research articles and four review articles that provide an introduction to several aspects of this rapidly developing field.Keywords: chemoinformatics; docking; force-field parameter; ligand-based virtual screening; molecular dynamics; molecular similarity; pharmacophore; structure-activity relationships; virtual screening Chemistry is, and has been for many years, one of the most information-rich academic disciplines. The very first journal devoted to chemistry was published as early as 1778, and the literature has grown steadily since then. Much of the information in chemistry relates to the structures-in both two and three dimensions-of individual chemical molecules; for example, the world's largest chemical database, the Chemical Registry produced by Chemical Abstracts Service, now contains the structures of over 90 million distinct molecules, and there are additional millions in other public databases and in the corporate files of pharmaceutical, agrochemical, and biotechnology companies. This wealth of information has spurred the development of a specialist discipline, that of chemoinformatics, which Gasteiger defines as "the application of informatics methods to solve chemical problems" [1]. Gasteiger has been one of the founding fathers of chemoinformatics, and his review included in this special issue provides a personal view of the history and development of the discipline over the past 50 years. As he notes, the novel modes of access that chemoinformatics provides to even the largest volumes of chemical information have profoundly affected the ways in which chemical research is conducted. That this is so is evidenced by the original research articles and review articles that comprise this special issue, since they provide an overview of current activity in this rapidly developing, and increasingly important field of chemistry.The structure of a molecule is a prime factor in determining its physical, chemical, and biological properties, and chemoinformatics draws on techniques from areas such as graph theory, multivariate statistics, and machine learning to provide sophisticated data mining facilities to correlate such properties with structure (however this is represented in computational form). Such correlation approaches lie at the heart of virtual screening, which is probably the most important current application of chemoinformatics. Virtual screening involves scanning a database-either of known molecules or of molecules that could in principle be made-to identify those with the greatest probabilities of exhibiting some property of interest, e.g., affecting an individual's cholesterol level or the viscosity of a lubricant. Several of the research articles in this issue disc...