LONG-TERM IDENTIFICATION OF STREPTOMYCETES USING PYROLYSIS MASS-SPECTROMETRY AND ARTIFICIAL NEURAL NETWORKS

Sixteen reference strains and thirteen fresh isolates of three putativ ely novel Streptomyces species were examined six times over twenty mon ths using pyrolysis mass spectrometry to examine the long-term reprodu cibility of the procedure. The reference strains and new isolates were correctly identified using information in each of the datasets and op erational fingerprinting, but direct statistical comparison of the dat asets for strain identification was unsuccessful between datasets. Art ificial neural networks were also used to identify the strains held in the datasets. Neural networks trained with pyrolysis mass spectra fro m a single dataset were found to successfully identify the reference s trains and fresh isolates in that dataset but were unable to identify many of the strains in the other datasets. However, a neural network t rained on representative pyrolysis mass spectra from each of the first three datasets were found to identify the reference strains and fresh isolates in those three datasets and in the three subsequent datasets . Therefore, artificial neural network analysis of pyrolysis mass spec trometric data can provide a rapid, cost-effective, accurate and long- term reproducible way of identifying and typing microorganisms.