Search

Graph clustering has received significant attention in recent years due to the breakthrough of graph neural networks (GNNs). However, GNNs frequently assume strong data homophily, which is not true in many real-world applications. Furthermore, practical graphs are typically noisy and sparse, which inevitably degrades the clustering performance. To this end, we propose a novel Contrastive Graph Clustering (CGC) method with adaptive filter framework. We first design an adaptive filter that can automatically learn a suitable filter for different data, mining holistic information beyond low-frequency components and encoding topology structure information into features. Afterward, we learn a refined graph based on a graph-level contrastive mechanism, which further boosts graph discriminability. Extensive experiments show that the proposed CGC method achieves significant improvement over state-of-the-art methods on several benchmark datasets. In particular, our simple method, which does not employ neural networks, outperforms many deep learning approaches. • High-frequency information is considered for graph clustering. • Contrastive regularizer is applied to refine graph structure. • Homophilic and heterophilic graphs are evaluated. [ABSTRACT FROM AUTHOR]

MALAT1 has been implicated in tumor progression. But the mechanism and role underlying MALAT1 in non-small cell lung cancer (NSCLC) cell resistance to gemcitabine (GEM) remain rarely understood. Through bioinformatics analysis, we predicted MALAT1/miR-27a-5p/PBOV1 regulatory axis and constructed GEM resistant A549/GEM cell line, and A549 was the parent cell line. qRT-PCR was utilized to assess MALAT1, miR-27a-5p and PBOV1 expression in A549 and A549/GEM cells. MTT method and colony formation assay were utilized to measure cell viability and cell proliferation. Flow cytometry was conducted to assess cell cycle and cell apoptosis. Wound healing and Transwell assays were conducted to measure cell migratory and invasive potentials. Dual-luciferase reporter gene assay and RNA immunoprecipitation were utilized to identify the targeted relationship between MALAT1 and miR-27a-5p, and the former assay was also utilized to determine the targeted relationship between miR-27a-5p and PBOV1. The impacts of MALAT1/miR-27a-5p/PBOV1 on tumor growth and GEM resistance of NSCLC cells in vivo were validated by using the tumor xenograft model. MALAT1 was observed to be highly expressed in tissues and cells of GEM resistant patients. Forced level of MALAT1 could markedly enhance A549 cell resistance to GEM, but this impact could be weakened by silencing MALAT1. MALAT1 downregulated miR-27a-5p level. PBOV1 was the target of miR-27a-5p and could significantly enhance GEM resistance of NSCLC cell. MALAT1 facilitated tumor growth in vivo via targeting miR-27a-5p/PBOV1 and enhanced resistance of NSCLC cells to GEM. MALAT1/miR-27a-5p/PBOV1 axis was implicated in NSCLC cell resistance to GEM. We deepened our understanding about how MALAT1 enhanced NSCLC cell resistance to GEM and provided development of therapeutic strategy for NSCLC with a possible target. • LncRNA MALAT1 was associated with gemcitabine resistance in NSCLC cells. • LncRNA MALAT1 targeted miR-27a-5p in NSCLC. • PBOV1 was the direct target of miR-27a-5p in NSCLC. • LncRNA MALAT1 enhanced non-small cell lung cancer cell resistance to gemcitabine by directly affecting miR-27a-5p/PBOV1 axis. [ABSTRACT FROM AUTHOR]

Recently, the all- d -metal Ni(Co)MnTi based Heusler compounds are found to have a giant magnetocaloric effect (GMCE) near room temperature and manifest different functionalities like multicaloric effects, which can be employed for solid-state refrigeration. However, in comparison to other traditional Heusler compounds, the relatively large thermal hysteresis (ΔT hys) and moderately steep ferromagnetic phase transition provides limitations for real applications. Here, we present that fast solidification (suction casting) can sufficiently tailor the GMCE performance by modifying the microstructure. Compared with the arc-melted sample, the magnetic entropy change of the suction-casted sample shows a 67% improvement from 18.4 to 29.4 Jkg−1K−1 for a field change (∆ μ 0 H) of 5 T. As the thermal hysteresis has maintained a low Δ T hys value (5.5 K) for the enhanced first-order phase transition, a very competitive reversible magnetic entropy change of 21.8 Jkg−1K−1 for ∆ μ 0 H = 5 T is obtained. Combining high-resolution transmission electron microscopy (HRTEM) and positron annihilation spectroscopy (PAS) results, the difference in lattice defect concentration is found to be responsible for the significant improvement in GMCE for the suction-cast sample, which suggests that defect engineering can be applied to control the GMCE. Our study reveals that fast solidification can effectively regulate the magnetocaloric properties of all- d -metal NiCoMnTi Heusler compounds without sacrificing Δ T hys. [Display omitted] [ABSTRACT FROM AUTHOR]

This work presented a novel net-shape-forming technology to prepare oxide dispersion strengthened (ODS) copper by cold spray additive manufacturing. There were three types of Al 2 O 3 particles in the internal oxidation Cu-Al 2 O 3 raw powders. After cold spraying, two types of nano-sized Al 2 O 3 inside the raw powder were inherited into the deposition, and the micro-sized Al 2 O 3 on the powder surface was crushed to nano-size under severe plastic deformation, which dispersed along the interface of deformed powder due to the plastic flow of Cu caused by adiabatic shearing. The plastic deformation produced by cold spraying was 20% higher than the hot extrusion with extrusion ratio of 11:1, which fully achieves the Al 2 O 3 fragmentation, dispersion distribution and densification of the cold sprayed deposition. Moreover, due to the significant work hardening and grain refinement effect by SPD during cold spraying, Vickers hardness and nanoindentation hardness of the ODS Cu deposition increased to 185.63 HV 3 and 2.76 GPa, respectively, which were significantly higher than the raw powders and hot extrusion sample. • The net-shape-forming of large-scale ODS Cu was fabricated by cold spray additive manufacturing. • ODS Cu deposition exhibited significant grain refinement and work hardening effect. • Nano-sized Al 2 O 3 particles inherited to the deposition. • Large-sized Al 2 O 3 particles broke due to the severe plastic deformation. [ABSTRACT FROM AUTHOR]