文摘
Secondary metabolites exhibit an astonishing multitude of functionalities and enormous chemical diversity and these qualities are responsible for their favoured selection as drug leads. The complex process of finding natural products’ bioactivities is largely based on trial and error, and is therefore risky, time- and cost-intensive. In recent decades, computer-assisted techniques have emerged as promising tools to manage the huge amount of available structural data of macromolecular targets and compounds annotated to specific functions, and to extract knowledge from these data for the prediction of new events. The novel concept of virtual parallel screening aims to access a pharmacological profile for each compound screened using an array of macromolecular targets. Providing putative ligand–target interactions, this in silico multitarget application meets the requirements for natural product research in a complementary way. It enables (i) a fast identification of potential targets (target fishing), (ii) insight into a putative molecular mechanism, and (iii) an estimation of the bioactivity profile which allows for prioritizing experimental investigations. The first application examples in natural product research are described.