Türk kuru incirinde mikotoksin riskinin sistematik olarak araştırılması ve tahmini

Fig (Ficus Carica L.) is a fruit of high nutritional value, with a significant food safety problem – mycotoxin, particularly aflatoxin (AF) contamination. Mycotoxin contamination in figs can be observed on the tree during pre-harvest and harvest, during processing, storage, and transportation. An in-depth investigation of the factors influencing mycotoxin contamination and predictive modeling studies are required for the development of effective control strategies and systems. In this thesis, The Rapid Alert System for Food and Feed (RASFF) database (2002-2020) was analyzed for mycotoxin hazard in Turkish dried figs. All the factors that can affect mycotoxin contamination (including weather conditions, agricultural and processing conditions, and non-product related factors (such as export quantities, regulatory changes)) were evaluated considering the yearly and monthly notification trends. Drought was identified as a key weather-related contributor to increased notifications in 2007 and 2012. Agriculture and process related factors that could have contributed to mycotoxin contamination in the investigated time frame were identified based on an in-depth review of academic and grey literature. A Recurrent Neural Network (RNN) based model was developed for the prediction of mycotoxin hazard (i.e. total number of RASFF notifications) from meteorological parameters for the primary dried fig producing regions in Turkey (İzmir and Aydın) (obtained from the Turkish General Directorate of Meteorology). The developed time series model predicted the total number of RASFF notifications with a r value of 0.64 and p<0.05. A web-based analysis and forecast tool for AF risk in Turkish dried figs containing two modules was designed based on the developed RNN model: Module 1 (designed for public use based on total number of RASFF notification data) and Module 2 (designed for use by the Republic of Türkiye Ministry of Agriculture and Forestry based on mycotoxin contamination data from the Ministry database). This tool has great potential as a comprehensive system for the minimization of the mycotoxin risk in dried figs. Systematic collection and analysis of data on mycotoxin contamination routes and influencing factors throughout, from pre-harvest to transportation coupled with the forecasting capability would provide a strong foundation for control efforts aimed at minimizing mycotoxin risk.