RAPID DETECTION OF VEROCYTOTOXIN PRODUCTION STATUS IN ESCHERICHIA-COLI BY ARTIFICIAL NEURAL-NETWORK ANALYSIS OF PYROLYSIS MASS-SPECTRA

Verocytotoxin (VT)-producing and VT-non-producing strains of Escherich ia coli of four different serogroups were characterised by pyrolysis-m ass spectrometry (Py-MS). Py-MS spectral data were used to train artif icial neural networks (ANNs) which then accurately assessed the VT-pro duction status of fresh clinical isolates of E. coli of the same serog roups from their Py-MS spectral data. Serogroup-specific ANNs could be trained successfully with Py-MS data from only one exemplar each of V T-producing and VT-non-producing strains and training was accomplished in less than 1 min. Where more than one VT-producing phenotype occurr ed within a serogroup it was necessary to include an example of each p henotype in the training set. The combination of Py-MS with ANNs may b e an important new, powerful and very rapid method for the detection o f a particular biological character, such as exotoxin production withi n strains of single species or sub-species of bacteria.