You are here:
PATTERN RECOGNITION/EXPERT SYSTEM FOR IDENTIFICATION OF TOXIC COMPOUNDS FROM LOW RESOLUTION MASS SPECTRA
Citation:
Scott, D. PATTERN RECOGNITION/EXPERT SYSTEM FOR IDENTIFICATION OF TOXIC COMPOUNDS FROM LOW RESOLUTION MASS SPECTRA. U.S. Environmental Protection Agency, Washington, D.C., EPA/600/J-95/275.
Description:
An empirical rule-based pattern recognition/expert system for classifying, estimating molecular weights and identifying low resolution mass spectra of toxic and other organic compounds has been developed and evaluated. he system was designed to accommodate low concentration spectra and provide some information for mixtures. t consists of a classifier followed by molecular weight estimators, filters and identification odules. Computer series of allowed molecular weights and selected base peaks for five classes are used in the filters to reduce misclassification and ensure correct identification. he target classes are nonhalobenzenes; chlorobenzenes; bromo- and bromochloro-alkanes/alkenes; mono- and dichloroalkanes/alkenes; and tri, tetra- and pentachloroalkanes/alkenes. he identification module for the 75 target compounds relies upon the high accuracy of the molecular weight estimators and base peak data for unique identification. he total system was extensively tested and reference spectra of 32 potential air pollutants, 99 randomly selected compounds, 37 gas chromatographic-mass spectroscopic (GC-MS) field spectra and with 400 pharmaceutical related spectra. ven with incomplete spectra the classification and identification performance was very good with accuracies of 97 (test, random and pharmaceutical) and 95% (field GC-MS). he median absolute deviations from the true molecular weights of the test, random, field and pharmaceutical spectra were 1-2 Da and the average absolute deviations were 6-10 Da. he program is very fast and runs on a personal computer.