Search

In federated learning (FL), malicious clients could manipulate the predictions of the trained model through backdoor attacks, posing a significant threat to the security of FL systems. Existing research primarily focuses on backdoor attacks and defenses within the generic federated learning scenario, where all clients collaborate to train a single global model. A recent study conducted by Qin et al. [24] marks the initial exploration of backdoor attacks within the personalized federated learning (pFL) scenario, where each client constructs a personalized model based on its local data. Notably, the study demonstrates that pFL methods with parameter decoupling can significantly enhance robustness against backdoor attacks. However, in this article, we whistleblow that pFL methods with parameter decoupling are still vulnerable to backdoor attacks. The resistance of pFL methods with parameter decoupling is attributed to the heterogeneous classifiers between malicious clients and benign counterparts. We analyze two direct causes of the heterogeneous classifiers: (1) data heterogeneity inherently exists among clients and (2) poisoning by malicious clients further exacerbates the data heterogeneity. To address these issues, we propose a two-pronged attack method, BapFL, which comprises two simple yet effective strategies: (1) poisoning only the feature encoder while keeping the classifier fixed and (2) diversifying the classifier through noise introduction to simulate that of the benign clients. Extensive experiments on three benchmark datasets under varying conditions demonstrate the effectiveness of our proposed attack. Additionally, we evaluate the effectiveness of six widely used defense methods and find that BapFL still poses a significant threat even in the presence of the best defense, Multi-Krum. We hope to inspire further research on attack and defense strategies in pFL scenarios. The code is available at: https://github.com/BapFL/code [ABSTRACT FROM AUTHOR]

Intramuscular fat (IMF) deposition is one of the most important factors affecting meat quality and is closely associated with the expression of carnitine palmitoyl transferase 1A (CPT1A) which facilitates the transfer of long-chain fatty acids (LCFAs) into the mitochondria. However, the role of how CPT1A regulates the IMF formation remains unclear. Herein, we established the temporal expression profile of CPT1A during the differentiation of goat intramuscular precursor adipocytes. Functionally, the knockdown of CPT1A by siRNA treatment significantly increased the mRNA expression of adipogenic genes and promoted lipid deposition in goat intramuscular precursor adipocytes. Meanwhile, a CPT1A deficiency inhibited cell proliferation and promoted cell apoptosis significantly. CPT1A was then supported by the overexpression of CPT1A which significantly suppressed the cellular triglyceride deposition and promoted cell proliferation although the cell apoptosis also was increased. For RNA sequencing, a total of 167 differential expression genes (DEGs), including 125 upregulated DEGs and 42 downregulated DEGs, were observed after the RNA silencing of CPT1A compared to the control, and were predicted to enrich in the focal adhesion pathway, cell cycle, apoptosis and the MAPK signaling pathway by KEGG analysis. Specifically, blocking the MAPK signaling pathway by a specific inhibitor (PD169316) rescued the promotion of cell proliferation in CPT1A overexpression adipocytes. In conclusion, the expression variation of CPT1A may reconstruct the lipid distribution between cellular triglyceride deposition and cell proliferation in goat intramuscular precursor adipocyte. Furthermore, we demonstrate that CPT1A promotes the proliferation of goat adipocytes through the MAPK signaling pathway. This work widened the genetic regulator networks of IMF formation and delivered theoretical support for improving meat quality from the aspect of IMF deposition. [ABSTRACT FROM AUTHOR]

Proteases play an essential role in the four sequential but overlapping phases of wound healing: hemostasis, inflammation, proliferation, and remodeling. In chronic wounds, excessive protease secretion damages the newly formed extracellular matrix, thereby delaying or preventing the normal healing process. Peptide-based fluorogenic sensors provide a visual platform to sense and analyze protease activity through changes in the fluorescence intensity. Here, we have developed an integrated microfluidic chip coated with multilayered fluorogenic nanofilms that can directly monitor protease activity. Fluorogenic protease sensors were chemically conjugated to polymer films coated on the surface of parallel microfluidic channels. Capillary flow layer-by-layer (CF-LbL) was used for film assembly and combined with subsequent sensor modification to establish a novel platform sensing technology. The benefits of our platform include facile fabrication and processing, controllable film nanostructure, small sample volume, and high sensitivity. We observed increased fluorescence of the LbL nanofilms when they were exposed to model recombinant proteases, confirming their responsiveness to protease activity. Increases in the nanofilms' fluorescence intensity were also observed during incubation with liquid extracted from murine infected wounds, demonstrating the potential of these films to provide real-time, in situ information about protease activity levels. [ABSTRACT FROM AUTHOR]

Smart surfaces that can dynamically respond to environmental stimuli have demonstrated great promise in wearable electronics and optical detectors. Herein, a photopatternable nanolayered polymeric film that can reversibly display and hide structural colors in the visible range in response to relative humidity (RH) changes is reported. This film is fabricated on a silicon substrate using layer‐by‐layer assembly of chitosan and photoreactive carboxymethyl cellulose‐azido derivative, and selectively crosslinked through UV irradiation from a photomask. Compared to crosslinked regions, un‐crosslinked ones swell more and result in a larger thickness at high RH; as a result, those two regions show distinguishable color differences. The correlation of displayed color with film thickness at various RHs is plotted in an International Commission on Illumination (CIE) chromaticity diagram, which is in good agreement with the results obtained from modeling based on the interference theory. It is demonstrated that the structural color patterns on the film can be hidden and displayed spontaneously by contact with humid air, including human breath. This humidity‐triggered color change is fast, of fine resolution, highly reversible, and compatible with most silicon‐based devices. This film is low‐cost, stable, and ready to be applied to large surface areas for potential applications in anticounterfeiting, humidity sensors, and optical color filters. [ABSTRACT FROM AUTHOR]