Artificial neural network-assisted colorimetric-photothermal dual-mode aptasensor for the detection of β-conglycinin.

Abstract

β-Conglycinin is the most important allergen in soy, and accurate detection of it is of great significance for ensuring food safety. In this study, a novel multifunctional bio-inspired nanozyme Au@His-MIL-88 was used as a signal probe to construct an artificial neural network (ANN) model-assisted colorimetric-photothermal dual-mode aptasensor by combining with catalytic hairpin assembly (CHA). The detection range of this method was 5-5000 ng/mL with a limit of detection (LOD) of 1.31 ng/mL. The recovery of β-conglycinin in real samples exhibited a range of 93.1-114.0 %, indicating its practicality. The integration of aptamers and CHA expanded the scope of target applications and improved detection sensitivity. Additionally, the ANN model can achieve deep integration and mutual correction of dual-mode sensing signals, thereby reducing regression errors and improving detection efficiency. The developed detection method has broad application prospects in the field of food allergen detection.