ABSTRACT The present study is undertaken to predict the effects of Zataria multflora essential oil, NaCl, temperature and pH on the probability percentage of growth of Clostridium botulinum using artificial neural network (ANN). To assess the effects of essential oil, NaCl, pH and temperature on Log P% of C. botulinum type A spores, the experiment was arranged in a factorial design. This design (4 ¥ 3 ¥ 3 ¥ 3) included four levels of EO (0.0, 0.015, 0.03 and 0.06%), three levels of NaCl (0.5, 4 and 6%), three levels of pH (6, 6.5 and 7.4), three storage temperatures (15, 25 and 35C) and repeated observations (12 times) for growth in BHI broth for up to 43 days (D). According to our results, the log P% of C. botulinum was affected significantly (P < 0.01) by Essential oil, pH, NaCl and temperature. The numbers of neurons in input hidden and output layers were 5, 11 and 1, respectively. The inputs to the ANN were temperature, pH, NaCl, essential oil and the number of days after inoculation. The output of the ANN was log P%. Our results show that the ANN is a suitable model for predicting the growth of C. botulinum when it is affected by different levels of pH, temperature, NaCl and essential oil. PRACTICAL APPLICATIONS Clostridium botulinum is an anaerobic bacterium that produces a powerful neurotoxin. Because there is concern over the safety of some chemical preservatives, an increased interest is started into more natural alternatives. Zataria multiflora has been used traditionally as flavor agent and its antimicrobial activities has been established in previous studies. In addition, predictive food microbiology involves the knowledge of microbial growth responses to environmental factors summarized as mathematical models. Over the last few years, artificial neural networks (ANNs), as nonlinear modeling techniques, have been proposed for use in predictive microbiology. Considering the importance of C. botulinum and using natural preservative in food safety and the power of ANN in predicting purposes, we suggest that ANN can be used as a prediction tool in food safety.

Keywords