Prediction of flavor of Maillard reaction product of beef tallow residue based on artificial neural network

Jingwei Cui; Yinhan Wang; Qiaojun Wang; Lixue Yang; Yiren Zhang; Emad Karrar; Hui Zhang; Qingzhe Jin; Gangcheng Wu; Xingguo Wang

doi:10.1016/j.fochx.2022.100447

Prediction of flavor of Maillard reaction product of beef tallow residue based on artificial neural network

Food Chem X. 2022 Sep 14:15:100447. doi: 10.1016/j.fochx.2022.100447. eCollection 2022 Oct 30.

Authors

Jingwei Cui^{1

2}, Yinhan Wang³, Qiaojun Wang⁴, Lixue Yang⁴, Yiren Zhang⁵, Emad Karrar^{1

2}, Hui Zhang^{1

2}, Qingzhe Jin^{1

2}, Gangcheng Wu^{1

2}, Xingguo Wang^{1

2}

Affiliations

¹ State Key Laboratory of Food Science and Technology, School of Food Science and Technology, National Engineering Research Center for Functional Food, International Joint Research Laboratory for Lipid Nutrition and Safety, Collaborative Innovation Center of Food Safety and Quality Control in Jiangsu Province, Jiangnan University, China.
² International Joint Laboratory on Food Safety, Jiangnan University, China.
³ Beijing Institute of Technology, School of Aerospace Engineering, China.
⁴ Guanghanshi Maidele Food CO., LTD, China.
⁵ Department of Chemistry, School of Physical Science, University of Liverpool, UK.

Abstract

The beef flavor of beef tallow residue was improved by enzymatic hydrolysis followed by the Maillard reaction, and the flavor could be predicted using an artificial neural network. Five beef tallow residue hydrolysates were prepared using different enzymes. The Flavourzyme and Papain (FP) hydrolysate had low molecular weight peptides and high degree of hydrolysis and free amino acid content. We identified 49 main compounds, including aldehydes, pyrazines, and furan. Furan and pyrazine were the dominant volatile compounds in the five beef tallow residue-derived Maillard reaction products (MRPs), and their profiles and levels in the FP MRPs were high. The FP MRPs had the best sensory characteristics. The artificial neural network analysis revealed that the multiple input single output model had a better performance than the single input single output model, and the prediction accuracy was>90%, indicating that the MRPs sensory evaluation scores could be accurately predicted.

Keywords: Artificial neural; Beef tallow residue-derived; Hydrolysis; Maillard reaction.