Library searches

Spectra and chemical structure searches are based on distance and similarity measures as introduced in Section 5.2. Different strategies are known sequential search, search based on inverted lists, and hierarchical search trees. The strategies are explained for search of spectra. 

This kind of search is based on comparing the measured spectrum with candidate spectra of the library, bit by bit. Sequential search is only useful if a small data set is to be treated or if it is obUgatory to retrieve every individual data set. A more efficient way is to sort the entities in a database by deriving appropriate keys. 

Every feature appears in the fist of keys together with the identity numbers (IDs) of all spectra that contain the actual feature. After collection of all features of the unknown spectrum, a rather short file can be generated on the basis of the keys that consist of all the candidate spectra. 

Problems may arise if the lengths of the inverted lists differ. This may be because certain wavenumbers are more typical than others or certain chemical structures appear more often, for example, -C-C- is more frequent than -C=C-. The solution to this problem is the appHcation of Hash coding algorithms the key is coded by a random number that is then stored in a random access file. 

Hierarchical arrangements of spectra or chemical structures are based on grouping of data by means of some similarity measure. 

The computer can hold a database where the main peaks of known products are stored. The spectra obtained by electron ionization alone are reproducible enough to be useful. 

Two types of library search have been developed. The first type, called forward search, compares the new spectrum with the spectra stored in the library and looks for the best match of the spectra. The second type, called reverse search, checks for the possible presence in the new spectra of a spectrum chosen in the library. 

FL01 1100,1001LIBRBRNES 14622 BENZOIC fiCID, 2-HYBB0XY-, 1-METHYLETHYL  

FL01 I100,1001LIBRfiKNBS SB66 BENZOIC ACID, 2-HYDROXY-pur 807 C7.H6.03 

Example of a library search. A spectrum of an aspirin sample B to D library spectra identified by the computer. 

---

T. L. Clerc, in Computer-Enhanced Analytical Spectroscopy, H. L. C. Meuzelaar, T.L. Isenhour (Eds.), Plenum Press, New York, 1987, pp. 145-162. Automated spectra interpretation and library search systems. 

Once a mass spectrum from an eluting component has been acquired, the next step is to try to identify the component either through the skill of the mass spectroscopist or by resorting to a library search. Most modem GC/MS systems with an attached data station include a large library of spectra from known compounds (e.g., the NIST library). There may be as many as 50,000 to 60,000 stored spectra covering most of the known simple volatile compounds likely to be met in analytical work. Using special search routines under the control of the computer, one can examine... 

Finally, note that the ions produced by the combined inlet and ion sources, such as electrospray, plasmaspray, and dynamic FAB, are normally molecular or quasi-molecular ions, and there is little or none of the fragmentation that is so useful for structural work and for identifying compounds through a library search. While production of only a single type of molecular ion may be useful for obtaining the relative molecular mass of a substance or for revealing the complexity of a mixture, it is often not useful when identification needs to be done, as with most general analyses. Therefore,... 

Once the peaks have been collected and stored, the computer can be asked to work on the data to produce a mass spectrum and print it out, or it can be asked to carry out other operations such as library searching, producing a mass chromatogram, and making an accurate mass measurement on each peak. Many other examples of the use of computers to process mass data are presented in other chapters of this book. 

However, the two levels may become obvious if the instrument operator tries, for example, to conduct a library search while the computer is trying to acquire input from another mass spectrum the library search has to wait. Acquiring the data is a foreground task. Other functions such as library searching are background tasks. 

The lesult of a gioup frequency identification can be used as a prefilter for library searches (57—59). 

Maximum benefit from Gas Chromatography and Mass Spectrometry will be obtained if the user is aware of the information contained in the book. That is, Part I should be read to gain a practical understanding of GC/MS technology. In Part II, the reader will discover the nature of the material contained in each chapter. GC conditions for separating specific compounds are found under the appropriate chapter headings. The compounds for each GC separation are listed in order of elution, but more important, conditions that are likely to separate similar compound types are shown. Part II also contains information on derivatization, as well as on mass spectral interpretation for derivatized and underivatized compounds. Part III, combined with information from a library search, provides a list of ion masses and neutral losses for interpreting unknown compounds. The appendices in Part IV contain a wealth of information of value to the practice of GC and MS. 

Figure 16.2 is the mass spectrum of propylene glycol and shows the presence of an abundant m/z 45 ion. A library search will provide strong evidence that this compound is propylene glycol. Preparation of a TMS derivative will confirm this assignment. 

There are many ways to interpret mass spectra. Frequently, prior knowledge or the results from a library search dictate the method. The proceeding is a brief description of an approach to mass spectral interpretation that is especially useful when little is known about the compounds in the sample. 

Even though a good fit is not obtained, the library search may indicate the structural type. Review the characteristic fragment pathways of the suspected structural type in Part II of this book, and check Part III to determine if the ions observed and neutral losses correspond to the suggested structural type. 

The mass spectra of drugs are as varied as the molecules from which they are formed (see Table 11.1). Two major sources that are available for identifying drugs are computer library search routines and Mass Spectral and GC Data of Drugs, Poisons and Their Metabolites. ... 

If you frequently analyze pesticides, obtain the latest edition of Mass Spectrometry of Pesticides and Pollutants (Safe and Hutzinger. Boca Raton, FL, CRC Press). This book, combined with the list of most abundant ions (Table 25.1) and/or a computer library search, will be sufficient to identify most commercial pesticides. Also, see Chapters 17, 26, and 27. 

Solvents and their impurities represent a wide class of compound types therefore, a discussion of common mass spectral features is meaningless. However, most of the mass spectra are listed in computer library search programs and The Eight Peak Index. ... 

If the belt moves too quickly, in relation to the rate of deposition, sample will not be deposited on all parts of the belt. This results in the production of an uneven total-ion-current (TIC) trace and a distortion of the mass spectra obtained, with consequent problems in interpretation, particularly if library searching is employed. 

Library searching The use of a computer to compare a mass spectrum to be identified with large numbers of reference spectra. 

H.J. Luinge, E.D. Ixussink, and T. Visser, Trace-level identity confirmation from infrared spectra by library searching and artificial neural networks. Anal. Chim. Acta, 345 (1997) 173-184. 

Table 5.9 summarises the main features of FTIR spectroscopy as applied to extracts (separated or not). Since many additives have quite different absorbance profiles FTIR is an excellent tool for recognition. Qualitative identification is relatively straightforward for the different classes of additives. Library searching entails a sequential, point-by-point, statistical correlation analysis of the unknown spectrum with each of the spectra in the library. Fully automated analysis of... 

A GC-IR-MS system with library search capability has been used to effectively identify the pyrolysis products of polybutadiene and the antioxidant additive 2,6-di-f-butyl-4-methylphenol [199]. Paper for food packaging was analysed by P T (at 100 °C) combined with /i-GC-UV. No specific applications of /rGC-UV to poly-mer/additive analysis have as yet been reported. 

---