Biomedical Engineering Reference
In-Depth Information
15
VIRTUAL SCREENING METHODS
J URGEN BAJORATH
15.1
INTRODUCTION
This chapter is an overview of current computational compound screening methods
[1]. Introductory sections are followed by a discussion of relevant methodologies.
Computational compound screening is commonly also referred to as in silico screen-
ing or virtual screening (VS). The latter term first appeared in the literature in 1997
[2] and is currently probably the most popular name. In VS, computational methods
are applied to screen large compound databases for novel active compounds (i.e., with
specific activity against a target of interest). As such, VS is often rationalized as a
complement to experimental high-throughput screening (HTS) [3-5]. In its essence,
VS is a hit identification approach, analogous to HTS. One might argue that VS
approaches might not really be required for practical purposes given the very large
HTS capacities that are available at present in the pharmaceutical industry. However,
this is clearly not the case. As has recently been pointed out [6], VS has produced a
considerable number of success stories, including the identification of first-in-class
inhibitors and novel active compounds in cases where HTS has not been successful.
Furthermore, VS frequently is a method of choice when assays are complex, time
consuming, and/or difficult to format for high throughput. Moreover, in academic
settings, where large HTS facilities are rarely available, VS is often the only feasible
approach to hit identification. However, VS can certainly not be expected to pro-
duce novel active compounds en masse. The majority of successful VS applications
currently yield limited numbers of hits that are predominantly weakly or moder-
ately potent, although there are exceptions [1,6-9]. The potency of HTS and VS
hits is often comparable, but HTS typically produces large numbers of hits because
 
Search WWH ::




Custom Search