The introduction and establishment of high-throughput screening (HTS) led to an increasing need for compounds to screen. Therefore, most pharmaceutical companies actively purchased compounds from various external sources and increased their own activities using combinatorial chemistry, a potential source of large numbers of new compounds for screening.

In the early days, combinatorial chemistry was characterized by experimenting with technology and methodology. The reactions had to be particularly robust and produce high yields; the selected reactants used in the synthesis of libraries were only allowed to carry functional groups that were compatible with the desired reaction or could be protected. In addition, they had to be rather rigid, cheap, and readily available. The limited selection or availability of reactants compatible with these restrictions often resulted in large libraries with a rather small number of different functional groups and a small range of physicochemical properties of the products such as octanol-water partition coefficients (log P) or molecular weight (MW). Large libraries with tens of thousands of compounds were synthesized and screened. The hit rate of these early, large libraries was very often disappointing. It became obvious that just increasing the number of structurally similar products did not usually increase the chance of interaction with biological targets. In addition, these large libraries often consisted of molecules with nondrug-like phys-icochemical properties such as high average molecular weight and lipophilicity [1].

Over the past years, computational methods have been used for the design of libraries with structurally diverse and more drug-like compounds as well as focused libraries that addressed a specific drug target [e.g. 2-6]. In particular, for focused libraries, in silico screening of virtual libraries increased the chance of synthesizing biologically active compounds and improving the screening hit rate [3, 7, 8]. In the following sections, we describe current methods of library design that are used for these purposes.

Handbook of Combinatorial Chemistry. Drugs, Catalysts, Materials. Vol. 2. Edited by K. C. Nicolaou, R. Hanko, and W. Hartwig Copyright © 2002 WILEY-VCH Verlag GmbH, Weinheim ISBN: 3-527-30509-2

762 | 27 Virtual Compound Libraries and Molecular Modeling 27.2

Was this article helpful?

0 0

Post a comment