Feasibility of Using Pyrolysis-Mass Spectrometry and Pyrolysis-MS/MS with Pattern Recognition for the Identification of Biological Materials

摘要

A series of experiments are reported which examined the susceptibility of the pyrolysis-mass spectroscopy (Py-MS) method in identifying bacteria which had been subjected to changes in growing time, killing method and diversity in the data set. In spite of these complications, Py-MS with pattern recognition techniques was successful where the number of bacteria in the data set was less than 16. The results from a large data set, containing 47 bacteria indicated that the number of samples exceeded the capabilities of the pattern recognition technique. Experiments using Py-MS/MS showed that this technique could provide enough selectivity to overcome the pattern recognition problems in large sample groups. Analyses were conducted on mixtures containing bacteria in a background which might be found on a battlefield. The complexity of the mixture and the level of the bacteria seriously affected the ability to identify the bacteria. Py-MS/MS with pattern recognition techniques will probably be necessary for future development of the technique.