KIAA0753 enhances osteoblast differentiation suppressed by diabetes.

Diabetes-related bone loss represents a significant complication that persistently jeopardizes the bone health of individuals with diabetes. Primary cilia proteins have been reported to play a vital role in regulating osteoblast differentiation in diabetes-related bone loss. However, the specific contribution of KIAA0753, a primary cilia protein, in bone loss induced by diabetes remains unclear. In this investigation, we elucidated the pivotal role of KIAA0753 as a promoter of osteoblast differentiation in diabetes. RNA sequencing demonstrated a marked downregulation of KIAA0753 expression in pro-bone MC3T3 cells exposed to a high glucose environment. Diabetes mouse models further validated the downregulation of KIAA0753 protein in the femur. Diabetes was observed to inhibit osteoblast differentiation in vitro, evidenced by downregulating the protein expression of OCN, OPN and ALP, decreasing primary cilia biosynthesis, and suppressing the Hedgehog signalling pathway. Knocking down KIAA0753 using shRNA methods was found to shorten primary cilia. Conversely, overexpression KIAA0753 rescued these changes. Additional insights indicated that KIAA0753 effectively restored osteoblast differentiation by directly interacting with SHH, OCN and Gli2, thereby activating the Hedgehog signalling pathway and mitigating the ubiquitination of Gli2 in diabetes. In summary, we report a negative regulatory relationship between KIAA0753 and diabetes-related bone loss. The clarification of KIAA0753's role offers valuable insights into the intricate mechanisms underlying diabetic bone complications.

Bifidobacterium infantis-Mediated Herpes Simplex Virus-TK/Ganciclovir Treatment Inhibits Cancer Metastasis in Mouse Model.

Previous studies have found that Bifidobacterium infantis-mediated herpes simplex virus-TK/ganciclovir (BF-TK/GCV) reduces the expression of VEGF and CD146, implying tumor metastasis inhibition. However, the mechanism by which BF-TK/GCV inhibits tumor metastasis is not fully studied. Here, we comprehensively identified and quantified protein expression profiling for the first time in gastric cancer (GC) cells MKN-45 upon BF-TK/GCV treatment using quantitative proteomics. A total of 159 and 72 differential expression proteins (DEPs) were significantly changed in the BF-TK/GCV/BF-TK and BF-TK/GCV/BF/GCV comparative analysis. Kyoto encyclopedia of genes and genomes (KEGG) pathway analysis enriched some metastasis-related pathways such as gap junction and cell adhesion molecules pathways. Moreover, the transwell assay proved that BF-TK/GCV inhibited the invasion and migration of tumor cells. Furthermore, immunohistochemistry (IHC) demonstrated that BF-TK/GCV reduced the expression of HIF-1α, mTOR, NF-κB1-p105, VCAM1, MMP13, CXCL12, ATG16, and CEBPB, which were associated with tumor metastasis. In summary, BF-TK/GCV inhibited tumor metastasis, which deepened and expanded the understanding of the antitumor mechanism of BF-TK/GCV.

PDGF-BB/PDGFRß promotes epithelial-mesenchymal transition by affecting PI3K/AKT/mTOR-driven aerobic glycolysis in Wilms' tumor G401 cells.

Wilms' tumor (WT) is the most common pediatric renal malignancy. PDGFRß belongs to the type III receptor tyrosine kinase family and is known to be involved in tumor metastasis and angiogenesis. Here, we studied the effect and underlying mechanism of PDGFRß on WT G401 cells. Transwell assay and wound-healing assay were used to detect the effect of PDGFRß on G401 cells invasion and migration. Western blot and immunofluorescence were used to detect the expression of EMT-related genes. The expression of PI3K/AKT/mTOR pathway proteins was detected by Western blot. The relationship between PDGFRß and aerobic glycolysis was studied by assessing the expression of glycolysis-related enzymes detected by qRT-PCR and Western blot. The activity of HK, PK, and LDH was detected by corresponding enzyme activity kits. The concentration of lactic acid and glucose was detected by Lactic Acid Assay Kit and Glucose Assay Kit-glucose oxidase method separately. To investigate the mechanism of PDGFRß in the development of WT, the changes of glucose and lactic acid were analyzed after blocking PI3K pathway, aerobic glycolysis, or PDGFRß. The key enzyme was screened by Western blot and glucose metabolism experiment after HK2, PKM2, and PDK1 were inhibited. The results showed that PDGFRß promoted the EMT process by modulating aerobic glycolysis through PI3K/AKT/mTOR pathway in which PKM2 plays a key role. Therefore, our study of the mechanism of PDGFRß in G401 cells provides a new target for the treatment of WT.

Alterations of brain activity and functional connectivity in transition from acute to chronic tinnitus.

The objective of this study was to investigate alterations to brain activity and functional connectivity in patients with tinnitus, exploring neural features in the transition from acute to chronic phantom perception. Twenty-four patients with acute tinnitus, 23 patients with chronic tinnitus, and 32 healthy controls were recruited. High-density electroencephalography (EEG) was used to explore changes in brain areas and functional connectivity in different groups. When compared with healthy subjects, acute tinnitus patients had a significant reduction in superior frontal cortex activity across all frequency bands, whereas chronic tinnitus patients had a significant reduction in the superior frontal cortex at beta 3 and gamma frequency bands as well as a significant increase in the inferior frontal cortex at delta-band and superior temporal cortex at alpha 1 frequency band. When compared to the chronic tinnitus group, the acute tinnitus group activity was significantly increased in the middle frontal and parietal gyrus at the gamma-band. Functional connectivity analysis showed that the chronic tinnitus group had increased connections between the parahippocampus gyrus, posterior cingulate cortex, and precuneus when compared with the healthy group. Alterations of local brain activity and connections between the parahippocampus gyrus and other nonauditory areas appeared in the transition from acute to chronic tinnitus. This indicates that the appearance and development of tinnitus is a dynamic process involving aberrant local neural activity and abnormal connectivity in multifunctional brain networks.

[Cellular primary cilia and human diseases].

Cellular primary cilium, located on the surface of virtually all mammalian cells, is a strictly conserved organelle which regulates cell biological process and maintains cell homeostasis by modulating cell proliferation, differentiation, migration, polarity, signal cascades and other life activities. Some diseases caused by mutations in genes encoding structural proteins or accessory proteins of primary cilia are collectively termed as "ciliopathies", which can occur in embryo, infancy and even adulthood. Ciliopathies not only involve a single organ, but also involve multiple organs and multiple systems, showing variable symptoms and overlapping symptoms. This review mainly summarizes the effects of ciliopathy-associated gene mutations on bone, tooth, skin, liver and bile duct, kidney, brain, retina, heart and other organs, uncovers their molecular mechanisms and provides some novel insights into therapy of ciliopathies.

Research on Bearing Fault Diagnosis Method Based on an Adaptive Anti-Noise Network under Long Time Series.

In the era of big data, longer time series fault signals will not only be easy to copy and store, but also reduce the labor cost of manual labeling, which can better meet the needs of industrial big data. Aiming to effectively extract the key classification information from a longer time series of bearing vibration signals and achieve high diagnostic accuracy under noise and different load conditions. The one-dimensional adaptive long sequence convolutional network (ALSCN) is proposed. ALSCN can better extract features directly from high-dimensional original signals without manually extracting features and relying on expert knowledge. By adding two improved multi-scale modules, ALSCN can not only extract important features efficiently from noise signals, but also alleviate the problem of losing key information due to continuous down-sampling. Moreover, a Bayesian optimization algorithm is constructed to automatically find the best combination of hyperparameters in ALSCN. Based on two bearing data sets, the model is compared with traditional model such as SVM and deep learning models such as convolutional neural networks (CNN) et al. The results prove that ALSCN has a higher diagnostic accuracy rate on 5120-dimensional sequences under -5 signal to noise ratio (SNR) with better generalization.

IFT80 Improves Invasion Ability in Gastric Cancer Cell Line via ift80/p75NGFR/MMP9 Signaling.

The assembly and maintenance of cilia depend on intraflagellar transport (IFT) proteins, which play an important role in development and homeostasis. IFT80 is a newly defined IFT protein and partial mutation of IFT80 in humans causes diseases such as Jeune asphyxiating thoracic dystrophy (JATD) and short rib polydactyly (SRP) type III, both characterized by abnormal skeletal development. However, the role and mechanism of IFT80 in the invasion of gastric cancer is unknown. We established SGC-7901 and MKN-45 gastric cancer cell lines that stably overexpressed IFT80, as verified by quantitative reverse transcription-PCR, Western blot, and immunofluorescence. Matrix metalloproteinase-9 (MMP9) plays an important role in tumor invasion, and its expression was assessed by quantitative reverse transcription-PCR, Western blotting, and immunofluorescence. The invasion ability of IFT80 on SGC-7901 and MKN-45 cells was examined by the Matrigel invasion assay. The relationship between p75NGFR, and the p75NGFR antagonists, PD90780 and IFT80, were detected by quantitative reverse transcription-PCR and Western blotting. We first detected an IFT80 expression pattern, and found that IFT80 was highly expressed in gastric cancer clinical samples. Overexpression of IFT80 in the gastric cancer cell lines, SGC-7901 and MKN-45, led to lengthening cilia. Additionally, overexpression of IFT80 significantly improved proliferation and invasion, but inhibited apoptosis, in gastric cancer cells. We further found that overexpression of IFT80 increased p75NGFR and MMP9 mRNA and protein expression. Treatment with the p75NGFR antagonist PD90780 inhibited the increased invasion ability resulting from overexpression of IFT80 in SGC-7901 and MKN-45 gastric cancer cells. Thus, these results suggest that IFT80 plays an important role in invasion of gastric cancer through regulating the ift80/p75NGFR/MMP9 signal pathways.

Bifidobacterial recombinant thymidine kinase-ganciclovir gene therapy system induces FasL and TNFR2 mediated antitumor apoptosis in solid tumors.

BACKGROUND: Directly targeting therapeutic suicide gene to a solid tumor is a hopeful approach for cancer gene therapy. Treatment of a solid tumor by an effective vector for a suicide gene remains a challenge. Given the lack of effective treatments, we constructed a bifidobacterial recombinant thymidine kinase (BF-rTK) -ganciclovir (GCV) targeting system (BKV) to meet this requirement and to explore antitumor mechanisms. METHODS: Bifidobacterium (BF) or BF-rTK was injected intratumorally with or without ganciclovir in a human colo320 intestinal xenograft tumor model. The tumor tissues were analyzed using apoptosis antibody arrays, real time PCR and western blot. The colo320 cell was analyzed by the gene silencing method. Autophagy and necroptosis were also detected in colo320 cell. Meanwhile, three human digestive system xenograft tumor models (colorectal cancer colo320, gastric cancer MKN-45 and liver cancer SSMC-7721) and a breast cancer (MDA-MB-231) model were employed to validate the universality of BF-rTK + GCV in solid tumor gene therapy. The survival rate was evaluated in three human cancer models after the BF-rTK + GCV intratumor treatment. The analysis of inflammatory markers (TNF-α) in tumor indicated that BF-rTK + GCV significantly inhibited TNF-α expression. RESULTS: The results suggested that BF-rTK + GCV induced tumor apoptosis without autophagy and necroptosis occurrence. The apoptosis was transduced by multiple signaling pathways mediated by FasL and TNFR2 and mainly activated the mitochondrial control of apoptosis via Bid and Bim, which was rescued by silencing Bid or/and Bim. However, BF + GCV only induced apoptosis via Fas/FasL signal pathway accompanied with increased P53 expression. We further found that BF-rTK + GCV inhibited the expression of the inflammatory maker of TNF-α. However, BF-rTK + GCV did not result in necroptosis and autophagy. CONCLUSIONS: BF-rTK + GCV induced tumor apoptosis mediated by FasL and TNFR2 through the mitochondrial control of apoptosis via Bid and Bim without inducing necroptosis and autophagy. Furthermore, BF-rTK + GCV showed to repress the inflammation of tumor through downregulating TNF-α expression. Survival analysis results of multiple cancer models confirmed that BF-rTK + GCV system has a wide field of application in solid tumor gene therapy.

Intravenous Administration Is an Effective and Safe Route for Cancer Gene Therapy Using the Bifidobacterium-Mediated Recombinant HSV-1 Thymidine Kinase and Ganciclovir.

The herpes simplex virus thymidine kinase/ganciclovir (HSV TK/GCV) system is one of the best studied cancer suicide gene therapy systems. Our previous study showed that caspase 3 expression was upregulated and bladder tumor growth was significantly reduced in rats treated with a combination of Bifidobacterium (BF) and HSV TK/GCV (BF-rTK/GCV). However, it was raised whether the BF-mediated recombinant thymidine kinase combined with ganciclovir (BF-rTK/GCV) was safe to administer via venous for cancer gene therapy. To answer this question, the antitumor effects of BF-rTK/GCV were mainly evaluated in a xenograft nude mouse model bearing MKN-45 gastric tumor cells. The immune response, including analysis of cytokine profiles, was analyzed to evaluate the safety of intramuscular and intravenous injection of BF-rTK in BALB/c mice. The results suggested that gastric tumor growth was significantly inhibited in vivo by BF-rTK/GCV. However, the BF-rTK/GCV had no effect on mouse body weight, indicating that the treatment was safe for the host. The results of cytokine profile analysis indicated that intravenous injection of a low dose of BF-rTK resulted in a weaker cytokine response than that obtained with intramuscular injection. Furthermore, immunohistochemical analysis showed that intravenous administration did not affect the expression of immune-associated TLR2 and TLR4. Finally, the BF-rTK/GCV inhibited vascular endothelial growth factor (VEGF) expression in mouse model, which is helpful for inhibiting of tumor angiogenesis. That meant intravenous administration of BF-rTK/GCV was an effective and safe way for cancer gene therapy.

Antimutagenic Effects of Selenium-Enriched Polysaccharides from Pyracantha fortuneana through Suppression of Cytochrome P450 1A Subfamily in the Mouse Liver.

Both selenium (Se) and polysaccharides from Pyracantha fortuneana (Maxim.) Li (PFPs) (P. fortuneana) have been reported to possess antioxidative and immuno-protective activities. Whether or not Se-containing polysaccharides (Se-PFPs) have synergistic effect of Se and polysaccharides on enhancing the antioxidant and immune activities remains to be determined. We previously reported that polysaccharides isolated from Se-enriched P. fortuneana (Se-PFPs) possessed hepatoprotective effects. However, it is not clear whether or not they have anti-mutagenic effects. In the present study, we compared and evaluated anti-mutagenic effects of Se-PFPs at three concentrations (1.35, 2.7 and 5.4 g/kg body weight) with those of PFPs, Se alone or Se + PFPs in mice using micronucleus assay in bone marrow and peripheral blood as well as mitomycin C-induced chromosomal aberrations in mouse testicular cells. We also elucidated the underlying mechanism. Our results demonstrated that Se-PFPs inhibited cyclophosphamide (CP)-induced micronucleus formation in both bone marrow and peripheral blood, enhanced the activities of superoxide dismutase (SOD) and glutathione peroxidase (GPx) in mouse liver, and reduced the activity and expression of cytochrome P450 1A (CYP4501A) in mouse liver in a dose-dependent manner. In addition, we found that the anti-mutagenic potential of Se-PFPs was higher than those of PFPs, Se alone or Se + PFPs at the same level. These results suggest that the anti-mutagenic potential of Se-PFPs may be mediated through the inhibition of the activity and expression of CYP4501A. This study indicates that application of Se-PFPs may provide an alternative strategy for cancer therapy by targeting CYP1A family.

The overexpression of MCPH1 inhibits cell growth through regulating cell cycle-related proteins and activating cytochrome c-caspase 3 signaling in cervical cancer.

MCPH1, initially identified as an hTERT repressor, has recently been implicated in mediating DNA damage response and maintaining chromosome integrity. This study is to investigate its potential role in the onset of cervical cancer. In the study, decreased expression of MCPH1 was observed in 19 of 31 cases (61.3%) at mRNA level and 44 of 63 cases (69.8%) at protein level of cervical tumor tissues compared with the paired nontumor tissues. Reduced MCPH1 protein expression was significantly associated with high-tumor grade (1 vs. 3 P = 0.013; 2 vs. 3 P = 0.047). In addition to inhibit SiHa cell migration and invasion, the overexpression of MCPH1 inhibited cervical cancer cells growth through inducing S phase arrest and mitochondrial apoptosis. Further analysis demonstrated cyclinA2/CDK2, CDC25C-cyclinB/CDC2, and p53/p21 pathways were involved in the MCPH1 overexpression-induced S phase arrest. Moreover, the overexpression of MCPH1 activated mitochondrial apoptosis through regulating several apoptosis-related proteins such as p53, Bcl-2, Bax, cytochrome c, caspase-3, and PARP-1. Our findings indicate that downregulated MCPH1 correlates with tumor progression in cervical cancer, and MCPH1 has an important role in regulating cell growth through regulating the cell cycle and apoptosis. Thus, it may be a crucial tumor suppressor gene and a novel candidate therapeutic target for cervical cancer.

Synthetic RGDS peptide attenuated lipopolysaccharide/D-galactosamine-induced fulminant hepatic failure in mice.

BACKGROUND AND AIM: Fulminant hepatic failure (FHF) is a serious clinic syndrome with extremely poor prognosis and no effective treatment except for liver transplantation. Synthetic RGDS peptide, an inhibitor of integrins, was proved to suppress integrin signals. In this study, we investigated the protection effects of RGDS peptide on lipopolysaccharide/D-galactosamine (LPS/D-GalN)-induced FHF and the underlying molecular mechanisms. METHODS: Synthetic RGDS peptide was given intraperitoneally 30 min before LPS/D-GalN injection. Liver function and the extent of liver injury were analyzed biochemically and pathologically respectively. Enzyme-linked immunosorbent assay, real-time polymerase chain reaction and Western blotting were used to detect effectors and signaling molecules. RESULTS: Pretreatment with synthetic RGDS peptide significantly improved LPS/D-GalN-induced mortality, and liver injury as determined by alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activities, as well as pathological analysis. In addition, RGDS peptide significantly reduced tumor necrosis factor (TNF)-α and macrophage inflammatory protein (MIP)-2 production, and decreased myeloperoxidase (MPO) and NF-κB activity. Furthermore, Western blotting indicated that the levels of phospho-integrin ß3, phospho-focal adhesion kinase (FAK) and phospho-p38 mitogen-activated protein kinases (MAPK) decreased with RGDS peptide pretreatment. CONCLUSION: Together, these data suggest that synthetic RGDS peptide protect against LPS/D-GalN-induced FHF by inhibiting inflammatory cells migration and blocking the integrin αVß3-FAK-p38 MAPK and NF-κB signaling.

IFT80 is essential for chondrocyte differentiation by regulating Hedgehog and Wnt signaling pathways.

Partial mutation of intraflagellar transport 80 (IFT80) in humans causes Jeune asphyxiating thoracic dystrophy (JATD) and short-rib polydactyly (SRP) syndrome type III. These diseases are autosomal recessive chondrodysplasias that share clinical similarities, including shortened long bones and constricted thoracic cage. However, the role and mechanism of IFT80 in the regulation of chondrocyte differentiation and function remain largely unknown. We hypothesize that IFT80 is required for the formation and function of cilia and plays a critical role in chondrogenic differentiation by regulating Hedgehog (Hh) and Wingless (Wnt) signaling pathways. To test this hypothesis, we first analyzed the IFT80 expression pattern and found that IFT80 was predominantly expressed in growth plate chondrocytes and during chondrogenic differentiation. Silencing IFT80 impaired cilia formation and chondrogenic differentiation in mouse bone marrow derived stromal cells (BMSCs), and decreased the expression of chondrocyte marker genes--collagen II and aggrecan. Additionally, silencing IFT80 down-regulated Hh signaling activity whereas up-regulated Wnt signaling activity. The overexpression of Gli2 in IFT80-silenced cells promoted chondrogenesis and recovered the chondrogenic deficiency from IFT80 silencing. Overall, our results demonstrate that IFT80 is essential for chondrocyte differentiation by regulating the Hh and Wnt signaling pathways.

Hepatoprotective effects of syringin on fulminant hepatic failure induced by D-galactosamine and lipopolysaccharide in mice.

The prognosis for fulminant hepatic failure (FHF) still remains extremely poor with a high mortality and, therefore, better treatments are urgently needed. Syringin, a main active substance isolated from Eleutherococcus senticosus, has been reported to exhibit immunomodulatory and anti-inflammatory properties. In this study, we investigated the effects and underlying mechanisms of syringin on lipopolysaccharide (LPS) and D-galactosamine (D-GalN)-induced FHF in mice. Mice were administered syringin (10, 30 and 100 mg kg(-1), respectively) intraperitoneally (i.p) 30 min before LPS/D-GalN then mortality and liver injury were evaluated subsequently. We found that syringin dose-dependently attenuated LPS/D-GalN-induced FHF, as indicated by reduced mortality, inhibited aminotransferase and malondialdehyde (MDA) content, an increased glutathione (GSH) concentration and alleviated pathological liver injury. In addition, syringin inhibited LPS/D-GalN-induced hepatic caspase-3 activation and hepatocellular apoptosis, myeloperoxidase (MPO) activity and intercellular adhesion molecule-1 (ICAM-1) expression, as well as hepatic tissues tumor necrosis factor-alpha (TNF-α) production and NF-κB activation in a dose-dependent manner. These experimental data indicate that syringin might alleviate the FHF induced by LPS/D-GalN through inhibiting NF-κB activation to reduce TNF-α production.

Minimally invasive percutaneous compression plating versus dynamic hip screw for intertrochanteric fractures: a randomized control trial.

OBJECTIVE: Intertrochanteric femur fracture is a common injury in elderly patients. The dynamic hip screw (DHS) has served as the standard choice for fixation; however it has several drawbacks. Studies of the percutaneous compression plate (PCCP) are still inconclusive in regards to its efficacy and safety. By comparing the two methods, we assessed their clinical therapeutic outcome. METHODS: A total of 121 elderly patients with intertrochanteric femur fractures (type AO/OTA 31.A1-A2, Evans type 1) were divided randomly into two groups undergoing either a minimally invasive PCCP procedure or a conventional DHS fixation. RESULTS: The mean operation duration was significantly shorter in the PCCP group (55.2 min versus 88.5 min, P<0.01). The blood loss was 156.5 ml±18.3 ml in the PCCP group and 513.2 ml±66.2 ml in the DHS group (P<0.01). Among the patients treated with PCCP, 3.1% needed blood transfusions, compared with 44.6% of those that had DHS surgery (P<0.01). The PCCP group displayed less postoperative complications (P<0.05). The mean American Society of Anesthesiologists score and Harris hip score in the PCCP group were better than those in the DHS group. There were no significant differences in the mean hospital stay, mortality rates, or fracture healing. CONCLUSION: Due to several advantages, PCCP has the potential to become the ideal choice for treating intertrochanteric fractures (type AO/OTA 31.A1-A2, Evans type 1), particularly in the elderly.

Tensorized Incomplete Multi-view Kernel Subspace Clustering.

Recently considerable advances have been achieved in the incomplete multi-view clustering (IMC) research. However, the current IMC works are often faced with three challenging issues. First, they mostly lack the ability to recover the nonlinear subspace structures in the multiple kernel spaces. Second, they usually neglect the high-order relationship in multiple representations. Third, they often have two or even more hyper-parameters and may not be practical for some real-world applications. To tackle these issues, we present a Tensorized Incomplete Multi-view Kernel Subspace Clustering (TIMKSC) approach. Specifically, by incorporating the kernel learning technique into an incomplete subspace clustering framework, our approach can robustly explore the latent subspace structure hidden in multiple views. Furthermore, we impute the incomplete kernel matrices and learn the low-rank tensor representations in a mutual enhancement manner. Notably, our approach can discover the underlying relationship among the observed and missing samples while capturing the high-order correlation to assist subspace clustering. To solve the proposed optimization model, we design a three-step algorithm to efficiently minimize the unified objective function, which only involves one hyper-parameter that requires tuning. Experiments on various benchmark datasets demonstrate the superiority of our approach. The source code and datasets are available at: https://www.researchgate.net/publication/381828300_TIMKSC_20240629.

Fine-Grained Learning Behavior-Oriented Knowledge Distillation for Graph Neural Networks.

Knowledge distillation (KD), as an effective compression technology, is used to reduce the resource consumption of graph neural networks (GNNs) and facilitate their deployment on resource-constrained devices. Numerous studies exist on GNN distillation, and however, the impacts of knowledge complexity and differences in learning behavior between teachers and students on distillation efficiency remain underexplored. We propose a KD method for fine-grained learning behavior (FLB), comprising two main components: feature knowledge decoupling (FKD) and teacher learning behavior guidance (TLBG). Specifically, FKD decouples the intermediate-layer features of the student network into two types: teacher-related features (TRFs) and downstream features (DFs), enhancing knowledge comprehension and learning efficiency by guiding the student to simultaneously focus on these features. TLBG maps the teacher model's learning behaviors to provide reliable guidance for correcting deviations in student learning. Extensive experiments across eight datasets and 12 baseline frameworks demonstrate that FLB significantly enhances the performance and robustness of student GNNs within the original framework.

Contrastive Multiview Attribute Graph Clustering With Adaptive Encoders.

Multiview attribute graph clustering aims to cluster nodes into disjoint categories by taking advantage of the multiview topological structures and the node attribute values. However, the existing works fail to explicitly discover the inherent relationships in multiview topological graph matrices while considering different properties between the graphs. Besides, they cannot well handle the sparse structure of some graphs in the learning procedure of graph embeddings. Therefore, in this article, we propose a novel contrastive multiview attribute graph clustering (CMAGC) with adaptive encoders method. Within this framework, the adaptive encoders concerning different properties of distinct topological graphs are chosen to integrate multiview attribute graph information by checking whether there exists high-order neighbor information or not. Meanwhile, the number of layers of the GCN encoders is selected according to the prior knowledge related to the characteristics of different topological graphs. In particular, the feature-level and cluster-level contrastive learning are conducted on the multiview soft assignment representations, where the union of the first-order neighbors from the corresponding graph pairs is regarded as the positive pairs for data augmentation and the sparse neighbor information problem in some graphs can be well dealt with. To the best of our knowledge, it is the first time to explicitly deal with the inherent relationships from the interview and intraview perspectives. Extensive experiments are conducted on several datasets to verify the superiority of the proposed CMAGC method compared with the state-of-the-art methods.

Interaction-knowledge semantic alignment for recommendation.

In order to alleviate the issue of data sparsity, knowledge graphs are introduced into recommender systems because they contain diverse information about items. The existing knowledge graph enhanced recommender systems utilize both user-item interaction data and knowledge graph, but those methods ignore the semantic difference between interaction data and knowledge graph. On the other hand, for the item representations obtained from two kinds of graph structure data respectively, the existing methods of fusing representations only consider the item representations themselves, without considering the personalized preference of users. In order to overcome the limitations mentioned above, we present a recommendation method named Interaction-Knowledge Semantic Alignment for Recommendation (IKSAR). By introducing a semantic alignment module, the semantic difference between the interaction bipartite graph and the knowledge graph is reduced. The representation of user is integrated during the fusion of representations of item, which improves the quality of the fused representation of item. To validate the efficacy of the proposed approach, we perform comprehensive experiments on three datasets. The experimental results demonstrate that the IKSAR is superior to the existing methods, showcasing notable improvement.

LKAN: LLM-Based Knowledge-Aware Attention Network for Clinical Staging of Liver Cancer.

Clinical staging of liver cancer (CSoLC), an important indicator for evaluating the degree of deterioration of primary liver cancer cells (PLCCs), is key in the diagnosis, treatment, and rehabilitation of liver cancer. In China, the current CSoLC adopts the China liver cancer (CNLC) staging, which is usually evaluated by clinicians based on the patient's radiology reports. Therefore, inferring clinical information from unstructured radiology reports can provide auxiliary decision support for clinicians. The key to solving the challenging task is to guide the model to pay attention to the staging-related words or sentences, and the following issues may occur: 1) Imbalanced categories: The symptoms of liver cancer in the early- or mid-stage are not obvious, resulting in more data in the end-stage. 2) Domain sensitivity of liver cancer data: The liver cancer dataset contains a large amount of domain knowledge, and the conventional methods can exacerbate out-of-vocabulary, which greatly affects the accuracy of classification. 3) Free-text and lengthy report: The radiology report of liver cancer sparsely describes various lesions with domain-specific terms, which poses difficulties in mining key information related to staging. To tackle these challenges, this article proposes a large language model (LLM)-based Knowledge-aware Attention Network (LKAN) for CSoLC. First, for maintaining semantic consistency, LLM and a rule-based algorithm are integrated to generate more diverse and reasonable data. Second, unlabeled radiology corpus of liver cancer are pre-trained to introduce domain knowledge for subsequent representation learning. Third, attention is improved by incorporating both global and local features, which can provide professional guidance for the classifier to focus on the important information. Compared with the baseline models, the classification accuracy of LKAN has achieved the best results with 90.3% Accuracy, 90.0% Macro_F1 score, and 90.0% Macro_Recall. The code is available at https://github.com/xczhh/Supplemental-Material.