Library optimization

With the advent of radars capable of waveform agility, the design of optimal waveform libraries comes into question. The purpose of this section is to consider the design of such waveform libraries for radar tracking applications, from an information theoretic point of view. We note that waveform libraries will depend in general on the specific applications in which the systems are to be used. Airborne radars will require different libraries from ship-borne ones. Radars used in a tracking mode will require different optimal libraries than radars in a surveillance mode. 

It was quickly realized that parallel synthesis techniques could be used to increase the efficiency of medicinal chemistry in this role [111]. Small optimization libraries were produced using parallel synthesis technologies and they took advantage of the ability... 

D. E. Enhancing the hit-to-lead properties of lead optimization libraries. 

Many library design methods require that the size (number of products) and configuration (numbers of reactants selected for each component) of the library are specified upfront. However, it is often difficult to determine optimum values a priori and usually there is a trade-off between these criteria and the other criteria to be optimized. Consider the design of a library where the aim is to maximize coverage of some cell-based chemistry space. It is clear that as more products are included in the library the chance of occupying more cells increases. Thus, an optimal library is likely to be one that represents a compromise in size and diversity. 

The receptor relevance of BCUT descriptors has inspired several groups to apply them in conjunction with other methods. Beno and Mason reported the use of simulated annealing to optimize library design using BCUT chemistry space and four-point pharmacophores concurrently (33) and the use of chemistry spaces in conjunction with property profiles (52). The application of such composite methods to target class library design is readily apparent. Pirard and Pickett reported the application of the chemometric method, partial least squares discriminant analysis, with BCUT descriptors to successfully classify ATP-site-directed kinase inhibitors active against five different protein kinases... 

Key words High-throughput chemistry, chemical library, random library, targeted library, optimization library, library design, biological activity, drug discovery. 

In the previous examples, random and targeted libraries are used to discover leads. Optimization libraries are employed when lead structures have already been identified, serving to improve the potency, selectivity, or other characteristics of the molecule. 

Fig. 1.13. Optimization library for human rhinovirus 3C protease (reprinted ( adapted or in part ) with permission from Journal of the American Chemical Society. Copyright 2001 American Chemical Society).

In this chapter, we will give a brief introduction to the basic concepts of chemoinformatics and their relevance to chemical library design. In Section 2, we will describe chemical representation, molecular data, and molecular data mining in computer we will introduce some of the chemoinformatics concepts such as molecular descriptors, chemical space, dimension reduction, similarity and diversity and we will review the most useful methods and applications of chemoinformatics, the quantitative structure-activity relationship (QSAR), the quantitative structure-property relationship (QSPR), multiobjective optimization, and virtual screening. In Section 3, we will outline some of the elements of library design and connect chemoinformatics tools, such as molecular similarity, molecular diversity, and multiple objective optimizations, with designing optimal libraries. Finally, we will put library design into perspective in Section 4. 

There are two major classes of libraries for drug discovery diverse libraries for lead discovery and focused libraries for lead optimization. Lead discovery libraries emphasize diversity while lead optimization libraries prefer similar compounds. The purpose of lead discovery libraries is to find lead matter and to provide potential active compounds for further optimization. Without any prior knowledge about the active compounds for a given target, it is reasonable to start with a library of enough chemical space coverage to demarcate the biologically relevant chemical... 

Did the candidate molecule arise from an optimization library of an existing lead or is it an analog of a lead structure discovered utilizing combinatorial methods ... 

This is the most intriguing aspect of the clinical candidate generated by the Houghten workers, since the discovery of HP228 was apparently the result of a lead generation library. This library provided the new chemical entity, which ultimately was the platform for further library optimization chemistry, which, in turn, provided the clinical candidate. This appears to be the only case to date of a lead generated by combinatorial methods, which, after optimization, has resulted in the clinical candidate. The other cases in the clinic that we have information about all arose from optimization libraries on existing leads. 

LY-334370 was synthesized in a lead optimization library. The original lead structure was accessed from a biased set of molecules related to a platform with an affinity for the receptor target. The clinical candidate was synthesized directly in the lead optimization library. 

Product-based selection is much more computationally demanding than reactant-based selection, however it has been shown that better optimized libraries can result [60, 61], especially when the aim is to optimize the properties of a library as a whole, such as diversity or the distribution of physicochemical properties. In addition, product-based selection is usually more appropriate for focused libraries which require consideration of the properties of the resulting products. 

A recent trend in library design is to optimize libraries over a number of properties simultaneously for example, whether a library is designed to be diverse, focused or some combination of the two, it is desirable that the library is cheap to synthesis and that the compounds contained within the library have drug-like physicochemical properties. Most approaches to multiobjective library design combine the different properties via a weighted-sum fitness function. For example, in the SELECT program the fitness function can have the following form ... 

Hoogenboom HR, Designing and optimizing library selection strategies for generating high-affinity antibodies, Trends Biotechnol., 15 62-70, 1997. 

---