Fingerprint Library Searching

The fact that mass, infra-red, NMR and ultra-violet spectra and, to a certain extent, also chromatographic retention data can be considered as molecular fingerprints , forms the basis of most computerised library search systems. Retrieval methods for characteristic chemical data and techniques for the comparison of human fingerprints have similar elements the first step is to clean up the raw data, then in many cases a data reduction is carried out by selection of prominent features. Finally, there is the comparison of unknown and reference data patterns, which, for a useful result, requires a statistical correlation to be established. In this paper no attention will be paid to feature selection. 

Factors 1 and 2 are correlated usually, a decrease/increase of the reproducibility of the fingerprint data will cause a decrease/increase of the specificity of that data. Factors 2 and 4 are also correlated. As a matter of fact, the reproducibility of the data involved in a search system is a crucial element of the design of the similarity index. Especially the interlaboratory reproducibility plays an important role, whenever multisource databases are being used. This reproducibility is determined by differences in samples, instruments, experimental conditions, performance of analysts and operators, introduction of coding errors, etc. As a consequence, a major factor determining the usefiilness of computer-aided library search systems is the extent to which the reproducibility of the relevant data is accounted for in the design of the similarity measure. 

Another major use of computers for HPLC data should be varietal identification (see later). Qualitatively, cereal storage proteins vary little within genotypes but significantly among different varieties, so they provide characteristic fingerprints. Varietal identification can be automated by computer comparisons with stored standard data. Scanlon et al. [80] showed that normalization of peak retention times provided sufficient precision for cultivar identification. Resulting data based on peak heights and times could be used in an automated library search to identify wheat varieties [81]. 

Spectral library searches of selected peaks in the fingerprint region indicated that the residue of the acetone extract might contain at least two types of surfactant materials. One was a PEG-3 C12-14 alcohol. The other material was a cocoampho-carboxypropionic acid, cocobetaine or oleyl betaine material, lauroyl sarcosine, or a tridecylsulfate lauramphocarboxyglycinate sodium salt. 

On the other hand, there is considerable interest to quantify the similarities between different molecules, in particular, in pharmacology [7], For instance, the search for a new drug may include a comparative analysis of an active molecule with a large molecular library by using combinatorial chemistry. A computational comparison based on the similarity of empirical data (structural parameters, molecular surfaces, thermodynamical data, etc.) is often used as a prescreening. Because the DFT reactivity descriptors measure intrinsic properties of a molecular moiety, they are in fact chemical fingerprints of molecules. These descriptors establish a useful scale of similarity between the members of a large molecular family (see in particular Chapter 15) [18-21],... 

Decornez et al. used a generalized kinase model and a combination of 2D (fingerprint based similarity) and 3D methods (docking) to develop a kinase family focused library (15). The authors used 2800 kinase inhibitors compounds as a reference for a 2D search of their in-house database of 260 compounds... 

Typically, dereplication is initiated with some analysis, chromatographic and/or spectroscopic, to recognise an active entity detected during HTS. Additional analyses are employed to rapidly establish the unambiguous identity of the compound. This fingerprint can then be used to search databases and reference libraries to link the structure to all chemical, spectral, bioactivity and pharmacokinetic data, as well as patent and publication information. 

More frequently than chemical techniques, the spectroscopic methods of analysis are used for the determination of polymer chemical composition. Among these techniques the use of infrared (IR) absorption spectra as fingerprints for polymer identification is probably the most common. The IR absorption is produced tjy the transition of the molecules from one vibrational quantum state into another, and most polymers generate characteristic spectra. Large databases containing polymer spectra (typically obtained using Fourier transform infra-red spectroscopy or FTIR) are available, and modern instruments have efficient search routines for polymer identification based on matching an unknown spectrum with those from the library. For specific polymers, the IR spectra can reveal even some subtle composition characteristics such as interactions between polymer molecules in polymeric blends. 

The unpredictability of what components will constitute a successful sensing layer underlines the power of utilizing a 2D combinatorial parallel approach to the discovery of successful sensing systems in aqueous media. The library response toward metal cations can be used to search for either a unique response (individual hit ) or a whole fingerprint of responses. Here, the fingerprint is the collection of the individual responses of each sensing layer to one cation. Rapid inspection of the library fingerprint (Fig. 4.11) provides a unique response for each cation. 

---