The problem of microbial biofilms has come to the fore alongside food, pharmaceutical, and healthcare industrialization. The development of new antibiofilm products has become urgent, but it includes bioprospecting and is time and money-consuming. Contemporary efforts are directed at the pursuit of effective compounds of natural origin, also known as "green" agents. Mushrooms appear to be a possible new source of antibiofilm compounds, as has been demonstrated recently. The existing modeling methods are directed toward predicting bacterial biofilm formation, not in the presence of antibiofilm materials. Moreover, the modeling is almost exclusively targeted at biofilms in healthcare, while modeling related to the food industry remains under-researched. The present study applied an Artificial Neural Network (ANN) model to analyze the anti-adhesion and anti-biofilm-forming effects of 40 extracts from 20 mushroom species against two very important food-borne bacterial species for food and food-related industries-Listeria monocytogenes and Salmonella enteritidis. The models developed in this study exhibited high prediction quality, as indicated by high r2 values during the training cycle. The best fit between the modeled and measured values was observed for the inhibition of adhesion. This study provides a valuable contribution to the field, supporting industrial settings during the initial stage of biofilm formation, when these communities are the most vulnerable, and promoting innovative and improved safety management.
Keywords: antiadhesion; antibiofilm; artificial neural network; food-borne pathogens; model; mushroom extracts.