Xihua University

Amplicon contamination remains a major challenge in nucleic acid biosensing, frequently compromising diagnostic accuracy in both clinical testing and field-deployable sensing systems. Although strategies such as uracil-DNA glycosylase (UDG) treatment and CRISPR-based approaches have been developed to mitigate this problem, their broader implementation is often limited by incomplete decontamination, increased assay cost, and the complexity associated with exogenous guide design. Here we introduce CREDIT (chopping-based Argonaute-mediated testing), a molecular diagnostic biosensor platform that harnesses the intrinsic chopping activity of Pyrococcus furiosus Argonaute (PfAgo). Unlike conventional Argonaute-based biosensing assays that primarily use Argonaute as a signal-generating enzyme, CREDIT operates as a dual-function system: it degrades amplification products into short fragments that are inert to further amplification, while simultaneously repurposing these chopping-derived fragments as guides to direct sequence-specific cleavage of reporter probes for signal generation. We demonstrate that this integrated detect-and-destroy mechanism enables seamless coupling of CREDIT with both PCR and isothermal amplification for ultrasensitive nucleic acid detection. The platform preserves the attomolar sensitivity of the amplification-based assay without compromising assay speed. Notably, a stringent 20-round consecutive open-lid stress test confirms complete suppression of amplicon contamination, thereby enabling uncapped post-amplification handling for lateral flow strip (LFS) sensor readout-an operation previously considered unsafe in field-deployable molecular testing. Together, these results establish CREDIT as a contamination-resistant biosensing strategy for molecular diagnostics that enables reliable and consecutive nucleic acid detection, particularly in resource-limited settings.

Transfer-based black-box adversarial attacks provide a practical means to evaluate the robustness of deep neural networks under restricted access to target models. However, existing approaches suffer from severe performance degradation in cross-architecture scenarios, particularly when adversarial examples crafted on Vision Transformers (ViTs) are transferred to convolutional neural networks (CNNs). We argue that this limitation cannot be fully attributed to optimization strategies alone, but may also be related to the mismatch between perturbation structure and model-specific inductive biases. To address this issue, we propose a structure-aware adversarial perturbation refinement framework that explicitly enforces spatial coherence during forward propagation. The proposed method consists of three components. First, spatial autocorrelation analysis is used to guide perturbation allocation toward structurally discriminative regions. Second, spatially connected perturbation patterns are introduced to help preserve perturbation effects under convolutional smoothing and pooling operations. Third, the perturbation strength is adaptively adjusted across network depth to balance structural disruption and semantic preservation. Extensive experiments on the ImageNet benchmark show that the proposed approach achieves competitive and often stronger performance than the compared transfer-based attacks across diverse ViT and CNN architectures, with particularly notable gains in the challenging ViT-to-CNN transfer setting. These results provide empirical support for the utility of structure-aware perturbation refinement in improving black-box adversarial transferability across heterogeneous visual architectures.

Glutathione S-transferase (GST) is a key enzyme involved in detoxification, antioxidation, and signal transduction, and its abnormal activity is closely associated with tumor drug resistance, liver damage, and inflammatory diseases. In this study, a bifunctional nanozyme based on novel iron-doped polymer dots (Fe-PDs) was constructed, which exhibits both yellow-green fluorescence and excellent oxidase-like activity. The Fe-PDs can oxidize N,N-diethyl-p-phenylenediamine (DPD) to generate a red oxidized product oxDPD (with an absorption peak at 551 nm), while simultaneously quenching their own fluorescence at 522 nm through fluorescence resonance energy transfer. Upon the addition of reduced glutathione (GSH), oxDPD is reduced, leading to a decrease in absorbance and restoration of fluorescence. Catalyzed by GST, GSH conjugates with 1-chloro-2,4-dinitrobenzene, consuming GSH in the reaction system and thereby re-triggering the oxidation of DPD, which enables dual-mode fluorescence and colorimetric detection. The fluorescence method shows a linear range of 0.001-5 U/L with a detection limit of 0.25 mU/L, while the colorimetric method achieves a linear range of 0.05-20 U/L with a detection limit of 13 mU/L. This platform shows promising potential for GST detection in human serum and for inhibitor screening. Furthermore, a portable hydrogel-based visual sensing platform was constructed, demonstrating good application performance in the analysis of GST in biological samples.