Deep learning assisted PfAgo-programmable genetic circuit for ultrasensitive visual detection of foodborne pathogen in one-tube.

Foodborne pathogens represent a significant threat to public health. The development of rapid and sensitive detection methods is critical for the effective prevention and control of food safety issues. In recent years, pathogens detection methods based on the Pyrococcus furiosus Argonaute (PfAgo) have received much attention. However, currently developed PfAgo-based pathogens detection methods still face problems such as long detection time and insufficient cleavage efficiency. Herein, we have developed an ultra-sensitive, visual assay for foodborne pathogens detection. Specifically, by designing PfAgo-based double-probe programmable genetic circuit to be compatible with the ultrafast thermal cycling V-shaped PCR (VPCR), we have not only enabled the process to be carried out in a one-tube, but also significantly improved sensitivity. In addition, we have developed a deep learning-based fluorescence image recognition technology (DL-FIR) to quickly, accurately, and in bulk process experimental results. This assay exceeds the detection performance of sequential cleavage in only one-third of the time and the fluorescence signal is amplified by over 200 %, significantly improving sensitivity (1 CFU/mL). This study presents a simple, sensitive, and universal assay for foodborne pathogens detection, enabling large-scale on-site screening during disease outbreaks.