Federated Graph Anomaly Detection via Contrastive Self-Supervised Learning.

Attribute graph anomaly detection aims to identify nodes that significantly deviate from the majority of normal nodes, and has received increasing attention due to the ubiquity and complexity of graph-structured data in various real-world scenarios. However, current mainstream anomaly detection methods are primarily designed for centralized settings, which may pose privacy leakage risks in certain sensitive situations. Although federated graph learning offers a promising solution by enabling collaborative model training in distributed systems while preserving data privacy, a practical challenge arises as each client typically possesses a limited amount of graph data. Consequently, naively applying federated graph learning directly to anomaly detection tasks in distributed environments may lead to suboptimal performance results. We propose a federated graph anomaly detection framework via contrastive self-supervised learning (CSSL) federated CSSL anomaly detection framework (FedCAD) to address these challenges. FedCAD updates anomaly node information between clients via federated learning (FL) interactions. First, FedCAD uses pseudo-label discovery to determine the anomaly node of the client preliminarily. Second, FedCAD employs a local anomaly neighbor embedding aggregation strategy. This strategy enables the current client to aggregate the neighbor embeddings of anomaly nodes from other clients, thereby amplifying the distinction between anomaly nodes and their neighbor nodes. Doing so effectively sharpens the contrast between positive and negative instance pairs within contrastive learning, thus enhancing the efficacy and precision of anomaly detection through such a learning paradigm. Finally, the efficiency of FedCAD is demonstrated by experimental results on four real graph datasets.

STUB1 increases adiponectin expression by inducing ubiquitination and degradation of NR2F2, thereby reducing hepatic stellate cell activation and alleviating non-alcoholic fatty liver disease.

BACKGROUND: Adiponectin (APN) has exhibited ameliorating effects on non-alcoholic fatty liver disease (NAFLD). This study investigates the roles of APN and its regulatory molecules in hepatic stellate cell (HSC) activation and the progression of NAFLD. METHODS: Mice were subjected to a high-fat diet (HFD) to establish NAFLD models. Liver tissue was examined for lipid metabolism, fibrosis, and inflammation. Mouse 3T3-L1 adipocytes were exposed to palmitic acid (PA) to mimic a high-fat environment. The conditioned medium (CM) from adipocytes was collected for the culture of isolated mouse HSCs. Gain- or loss-of-function studies of APN, nuclear receptor subfamily 2 group F member 2 (NR2F2), and STIP1 homology and U-box containing protein 1 (STUB1) were performed to analyze their roles in NAFLD and HSC activation in vivo and in vitro. RESULTS: APN expression was poorly expressed in HFD-fed mice and PA-treated 3T3-L1 adipocytes, which was attributed to the transcription inhibition mediated by NR2F2. Silencing of NR2F2 restored the APN expression, ameliorating liver steatosis, fibrosis, and inflammatory cytokine infiltration in mouse livers and reducing HSC activation. Similarly, the NR2F2 silencing condition reduced HSC activation in vitro. However, these effects were counteracted by artificial APN silencing. STUB1 facilitated the ubiquitination and protein degradation of NR2F2, and its upregulation mitigated NAFLD-like symptoms in mice and HSC activation, effects reversed by the NR2F2 overexpression. CONCLUSION: This study highlights the role of STUB1 in reducing HSC activation and alleviating NAFLD by attenuating NR2F2-mediated transcriptional repression of APN.

Based on molecular structures: Amyloid-ß generation, clearance, toxicity and therapeutic strategies.

Amyloid-ß (Aß) has long been considered as one of the most important pathogenic factors in Alzheimer's disease (AD), but the specific pathogenic mechanism of Aß is still not completely understood. In recent years, the development of structural biology technology has led to new understandings about Aß molecular structures, Aß generation and clearance from the brain and peripheral tissues, and its pathological toxicity. The purpose of the review is to discuss Aß metabolism and toxicity, and the therapeutic strategy of AD based on the latest progress in molecular structures of Aß. The Aß structure at the atomic level has been analyzed, which provides a new and refined perspective to comprehend the role of Aß in AD and to formulate therapeutic strategies of AD.

Highly multiplexed quantifications of 299 somatic mutations in colorectal cancer patients by automated MALDI-TOF mass spectrometry.

BACKGROUND: Detection of somatic mutations in tumor tissues helps to understand tumor biology and guide treatment selection. Methods such as quantitative PCR can analyze a few mutations with high efficiency, while next generation sequencing (NGS) based methods can analyze hundreds to thousands of mutations. However, there is a lack of cost-effective method for quantitatively analyzing tens to a few hundred mutations of potential biological and clinical significance. METHODS: Through a comprehensive database and literature review we selected 299 mutations associated with colorectal cancer. We then designed a highly multiplexed assay panel (8-wells covering 299 mutations in 109 genes) based on an automated MADLI-TOF mass spectrometry (MS) platform. The multiplex panel was tested with a total of 319 freshly frozen tissues and 92 FFPE samples from 229 colorectal cancer patients, with 13 samples also analyzed by a targeted NGS method covering 532 genes. RESULTS: Multiplex somatic mutation panel based on MALDI-TOF MS detected and quantified at least one somatic mutation in 142 patients, with KRAS, TP53 and APC being the most frequently mutated genes. Extensive validation by both capillary sequencing and targeted NGS demonstrated high accuracy of the multiplex MS assay. Out of 35 mutations tested with plasmid constructs, sensitivities of 5 and 10% mutant allele frequency were achieved for 19 and 16 mutations, respectively. CONCLUSIONS: Automated MALDI-TOF MS offers an efficient and cost-effective platform for highly multiplexed quantitation of 299 somatic mutations, which may be useful in studying the biological and clinical significance of somatic mutations with large numbers of cancer tissues.

A Closed-Tube Nested Quantitative PCR Assay for Rapid Detection of Intron 22 Inversions in the Factor VIII Gene.

BACKGROUND: An inversion of intron 22 in the Factor VIII gene (Inv22) is the causative mutation for 45% of severe hemophilia A cases. Available methods for molecular diagnosis of Inv22 are generally tedious and not ideal for routine clinical use. METHODS: We report here a new method using a single closed-tube nested quantitative PCR (CN-qPCR) for rapid detection of Inv22. This method combines a 12-cycle long-distance PCR (LD-PCR) amplifying the int22h regions, followed by a duplex qPCR targeting two specific regions close to the int22h regions. All reagents were added to a single PCR mixture for the closed-tube assay. Sequential LD-PCR and qPCR was achieved by designing primers at substantially different melting temperatures and optimizing PCR conditions. RESULTS: Seventy-nine male hemophilia A patients of different disease severity were tested by both the CN-qPCR assay and the standard LD-PCR assay. CN-qPCR successfully made calls for all samples, whereas LD-PCR failed in eight samples. For the 71 samples where both methods made calls, the concordance was 100%. Inv22 was detected in 17 out of the 79 samples. Additionally, CN-qPCR achieved clear separation for 10 female carriers and 10 non-Inv22 females, suggesting the assay may also be useful for molecular diagnosis of female carriers. CONCLUSIONS: This new CN-qPCR method may provide a convenient and accurate F8 Inv22 test suitable for clinical use.