Library computer-based searches

POLYMERS, ADDITIVES, SURFACTANTS AND THEIR MIXTURES COMPUTER-BASED SEARCH WITH SPECIAL FTIR LIBRARIES... 

Sadtler offer nearly 200 000 digital infrared reference spectra in over fifty different collections and also publish handbooks and guides which cover the areas mentioned above. The Sadtler computer-based search system" and the other systems available from manufacturers such as Nicolet, Perkin Elmer, Bio Rad, etc., are all relatively easy to use. Sadtler also offer a computer-based system which contains both IR and NMR data, etc. Library search software packages, such as the Sadtler IR SearchMaster Software, the Spectrafile IR Search Software or the Spectra Calc Search Software, are frequently offered by FT-IR manufacturers in addition to specific search software formatted to operate with their particular data-stations/instruments/computer systems. Some of these search facilities may also cover a number of libraries not only of different suppliers but also of other techniques such as UV, NMR, MS etc. Obviously, such search software packages are dependent not only on the instrument but also on the user s interests. It should be home in mind that the information retrieved from some search software may not cover certain aspects which may normally be available from the particular Sadtler library being searched, such as physical properties, molecular structure. Chemical Abstracts Service (CAS) Registry Number, common impurities, etc. 

Gas Chromatography-Mass Spectrometry. A Finnigan 4500/Incos instrument with a 30-m X 0.32-mm i.d. capillary column coated with SP-B-5 was used. The GC parameters were as follows injector, 270 °C column oven temperature programmed, 50 °C (0.1 min, hold) 15 °C/min to 100 °C, 5 °C/min to 270 °C internal standard, anthracene-djo helium flow, 3.0 mL/min sample size, 3.0 /xL. MS conditions were as follows El, 70 eV scan (m/z), 35-650 daltons source temperature, 250 °C filament current, 0.5 A sensitivity, 10-8 A/V. (NOTE When the name of a compound is followed by (confirmed) , it means that the standard material was analyzed for confirmation under conditions identical to those of the sample when the name is followed by (tentative) , it means that the mass fragmentography showed the best fit (>80 ) based on the National Bureau of Standards [NBS] library computer search.)... 

In 1981 a new list with 86 additional substances became available (ref. 8). The computer search of the RTECS was reasonably straightforward at that time. The key word "teratogen" was used and the RTECS data base searched for the names of compounds under this classification. The size of the printed file was reasonable (500-600 entries). A copy of the file could be posted in the laboratory, with a short explanation that the list of teratogens should be viewed with caution due to the fact that data from the RTECS are unevaluated and that uncertainties exist about animal-human extrapolations. Any librarian having access to NLM-MEDLARS (National Library of Medicine-... 

The third key issue involves data handling. To take full advantage of the complementary nature of the data, it is desirable to handle all the spectral information in a single system. In our lab the IR and MS data are handled separately by the Nicolet and Hewlett Packard computer systems, respectively. Current commercial instruments also handle MS and IR data on separate systems, with only limited communications between the two computers. On our instrument, the following computer-based libraries were used to search unknown spectra IR - EPA Vapor Phase MS - National Bureau of Standards, Revision F. 

The data acquisition and processing operations on the MIKES instrument are under computer control and we are beginning to accumulate a library of reference spectra. A computer based library search routine has not yet been implemented on the MIKES instrument but existing gc/ms software can readily be adapted to the ms/ms data base. The construction of a ms/ms data base is underway in several laboratories. 

In addition to a library computer search for matching spectra, the usual procedure for interpretation of a mass spectrum involves (1) identification of the molecular ion peak and, if possible, determination of the elemental composition of the molecule based on the observed mass number (m/z) for the molecular ion, (2) establishment of the fragmentation patterns of the molecular ion, and finally (3) reduction of the number of possible structures for the compound, based on the interrelationships of the observed fragmentation patterns. The absence of meta-stable ions mass spectra obtained by GC-MS analysis makes (2) very difficult to achieve. 

Another place where one can submit SMILES or molhles and have a similarity score computed is described at http // www.jchem.com/doc/user/Compr.html (library comparison based on similarity) and at http //www.jchem.com/doc/ nser/Jcsearch.html (for all kinds of searching inclnding the Tanimoto-based similarity method). Both of these methods are inclnded in the JChem software (http //www.chemaxon. com/prodncts.html), which is free for academic nse. 

The information in Table 4.1 indicates that the unknown liquid is most likely l-chloro-2-methylbenzene since the probability of a correct match is placed at 94%. It is interesting to note that the meta and para isomers show probabilities of 70% and 60%, respectively. It is tempting to simply accept the results of the computer-based library search as correct, but the method is not an absolute guarantee that the identity of a sample has been correctly determined. A visual inspection of the experimental and library spectra must be included as part of the process. A computer can compare a mass spectrum it has determined with the spectra in these databases. 

The Library List window at the top right of Figure 8.5 allows the user to select the mass spectra libraries to be searched. The Min Pre-Search Hits window fixes the minimum number of spectra that will be selected at the end of the pre-research phase (see above). Increasing this value expands the length of the research (a small limitation based on the speed of modem computers) but reduces the risks of nnissing a solution if it exists. 

In view of latter developments (see Sections 16.4.9-16.4.11 for further details) the procedure, even with simplifications such as using a single CONCORD/ CORINA-derived 3D geometry instead of performing a Monte Carlo conformational search, is too computationally expensive to be applied to e-screening of virtual libraries. However, it may still be a useful alternative/complement for computing more detailed information about a compound, or to provide a more easily interpretable model to complement other models based on more rapidly computable parameters but which are difficult to interpret in terms of how to modify compounds in order for them to have better intestinal absorption characteristics. 

---