Search

Most existing multi-view graph clustering models either seek consistent clustering results from similarity matrices and spectral embeddings respectively or follow direct bidirectional integration of them, which ignores the interaction between them. To make up for this flaw, this paper designs a novel multi-view clustering model that performs Bidirectional Fusion with Cross-view Graph Filter (BF-CGF). To be specific, BF-CGF first learns a consistent graph embedding via performing the interaction between multi-view graphs and spectral embeddings with the perspective of the graph spectral domain and then considers seeking a consistent indicator matrix via the graph cut model from the consistent graph embedding and the similarity matrices. To solve the optimization problem of BF-CGF, we propose an efficient iterative algorithm and provide the corresponding convergence and complexity analyses. Extensive experimental results demonstrate that the proposed BF-CGF outperforms state-of-the-art competitors in most benchmark datasets.

Recently, graph-based multi-view clustering (GMC) has attracted extensive attention from researchers, in which multi-view clustering based on structured graph learning (SGL) can be considered as one of the most interesting branches, achieving promising performance. However, most of the existing SGL methods suffer from sparse graphs lacking useful information, which normally appears in practice. To alleviate this problem, we propose a novel multi-view and multi-order SGL ( $\text{M}^{2}$ SGL) model which introduces multiple different orders (multi-order) graphs into the SGL procedure reasonably. To be more specific, $\text{M}^{2}$ SGL designs a two-layer weighted-learning mechanism, in which the first layer truncatedly selects part of views in different orders to retain the most useful information, and the second layer assigns smooth weights into retained multi-order graphs to fuse them attentively. Moreover, an iterative optimization algorithm is derived to solve the optimization problem involved in $\text{M}^{2}$ SGL, and the corresponding theoretical analyses are provided. In experiments, extensive empirical results demonstrate that the proposed $\text{M}^{2}$ SGL model achieves the state-of-the-art performance in several benchmarks.

Existing multilayer graph clustering models focus on integrating the monolithic structure of each layer, but the local preference between layers and clusters has not been fully exploited. To alleviate this problem, this paper proposes a novel multilayer graph clustering model with Adaptive Local Modularity Learning (ALML), which mines truncated layer-cluster relationships adaptively with graph modularity learning paradigm, to reveal the local preference in multilayer graphs. More importantly, ALML reveals the equivalence between graph cut and graph modularity learning in multilayer graph scenarios in theory. To solve the optimization problem involved in ALML, this paper proposes an efficient alternating algorithm with quadratic-level time complexity, which is satisfactory in multilayer graph clustering scenarios, and provides corresponding analyses. In the simulations, extensive experimental results demonstrate that ALML outperforms state-of-the-art competitors in terms of effectiveness as well as efficiency.

In this issue of Chem Catalysis, Wang et al. have synthesized a single-atom Au3Fe1/Mo alloy featuring medium-spin Fe(III) through an alloying strategy to enhance plasmonic nitrogen fixation. The enhancement mechanism has been found to originate from the medium-spin Fe centers acting as active sites that facilitate the adsorption and activation of nitrogen molecules.

The cobalt-based catalysts were broadly investigated in NO reduction with CO; however, further development is limited by the poor low-temperature catalytic activity and H2O/O2resistance. Herein, a series of Ag-doped Co3O4porous nanosheets (xAgCo) were synthesized and applied into NO reduction with CO reaction. The formation of more oxygen vacancies of xAgCo catalysts was promoted by the weakened Co–O bond strength; therefore, the redox properties and oxygen mobility were enhanced. The xAgCo catalysts reached superior catalytic activities compared with Co3O4. Especially, 0.02AgCo exhibited the optimal catalytic performance, achieving the lowest apparent active energy (Ea= 34 kJ mol–1). The enhanced catalytic activity of xAgCo could be attributed to the optimized CO adsorption, promoted NO dissociation, and the formation of *NCO species. The 5 vol % H2O enhanced the NO conversion through promoting the decomposition of accumulated carbonate species. Meanwhile, the O2or 10 vol % H2O inhibited the NO conversion due to the coverage of active sites and the competitive adsorption/reaction.

A novel cascade strategy is proposed to construct Pd-NiCo2O4electrocatalyst.First time discovery that Ni incorporation together with Pd loading is able to balance the competitive adsorption of OH–and 5-hydroxymethylfurfural.Pd–NiCo2O4 promotes both the indirect and direct synergistic oxidation process.Pd–NiCo2O4catalyst exhibits extraordinary current density and excellent Faradaic Efficiency at a low potential.

Identifying drug–protein interactions (DPIs) is crucial in drug discovery, and a number of machine learning methods have been developed to predict DPIs. Existing methods usually use unrealistic data sets with hidden bias, which will limit the accuracy of virtual screening methods. Meanwhile, most DPI prediction methods pay more attention to molecular representation but lack effective research on protein representation and high-level associations between different instances. To this end, we present the novel structure-aware multimodal deep DPI prediction model, STAMP-DPI, which was trained on a curated industry-scale benchmark data set. We built a high-quality benchmark data set named GalaxyDB for DPI prediction. This industry-scale data set along with an unbiased training procedure resulted in a more robust benchmark study. For informative protein representation, we constructed a structure-aware graph neural network method from the protein sequence by combining predicted contact maps and graph neural networks. Through further integration of structure-based representation and high-level pretrained embeddings for molecules and proteins, our model effectively captures the feature representation of the interactions between them. As a result, STAMP-DPI outperformed state-of-the-art DPI prediction methods by decreasing 7.00% mean square error (MSE) in the Davis data set and improving 8.89% area under the curve (AUC) in the GalaxyDB data set. Moreover, our model is an interpretable model with the transformer-based interaction mechanism, which can accurately reveal the binding sites between molecules and proteins.

The mortality rate of esophageal squamous cell carcinoma (ESCC) is higher than that of other cancers worldwide owing to a lack of therapeutic targets and related drugs. This study aimed to find new drugs by targeting an efficacious therapeutic target in ESCC patients. Signal transducer and activator of transcription 3 (STAT3) is hyperactive in ESCC. Herein, we identified a novel STAT3 inhibitor, periplogenin, which strongly inhibited phosphorylation of STAT3 at Tyr705. Docking models and pull-down assays revealed that periplogenin bound directly and specifically to STAT3, leading to significant suppression of subsequent dimerization, nuclear import, and transcription activities. In addition, STAT3 knockdown cell lines were insensitive to periplogenin, whereas in contrast, STAT3-overexpressing cells were more sensitive to periplogenin, indicating that STAT3 was a target of periplogenin. Intraperitoneally administered periplogenin exhibited efficacious therapeutic effects in ESCC patient-derived xenograft models and dramatically impaired the phosphorylation of STAT3 and expression levels of STAT3-mediated downstream genes. Thus, our study demonstrated that periplogenin acted as a new STAT3 inhibitor, suppressing the growth of ESCC in vitro and in vivo, providing a basis for its potential application in ESCC treatment and prevention.

Objectives We assessed the efficacy of a 3-week primary or salvage caspofungin regimen in patients with chronic obstructive pulmonary disease (COPD) and concomitant proven or suspected invasive pulmonary aspergillosis (IPA).Methods Forty-four patients were treated with an initial loading caspofungin dose of 70 mg, followed by a daily dose of 50 mg for 20 days. The main efficacy endpoint was clinical effectiveness. Secondary endpoints included the clinical efficacy of caspofungin after 1 week, therapeutic efficacy based on the European Organization for Research and Treatment of Cancer and Mycoses Study Group Education and Research Consortium (EORTC/MSG) criteria, the sensitivity of different Aspergillusstrains to caspofungin in vitro, and the safety of caspofungin.Results An assessment of 42 patients in the intention-to-treat group revealed efficacy rates of 33.33% within 1 week and 38.10% within 3 weeks. According to the EORTC/MSG criteria, the treatment success rate was 38.10%. The success rate of first-line treatment was 54.76%, whereas salvage treatment had a success rate of 45.24%. No adverse events were reported among the participants.Conclusions Caspofungin is effective and safe as an initial or salvage treatment for patients with IPA and COPD.

Electrochemical dechlorination of 1,2-dichloroethane (DCE) is one of the prospective and economic strategies for the preparation of high-value ethylene. However, the exploration of advanced electrocatalysts with high reactivity and selectivity and the identification of their active sites are still a challenge. Herein, a single-atom (SA) Fe–Nx–C nanosheet with the presence of a highly efficient Fe–N4coordination pattern is reported. The as-prepared single-atom electrocatalyst exhibits a higher reactivity and ethylene selectivity for DCE dechlorination reaction than those of the commercially adopted 20% Pt–C catalyst. By a combination of experiments and theoretical calculations, the atomically dispersed Fe center in the Fe–N4structure was unveiled to be the dominating active site for electrochemical production of ethylene. Our work would offer an approach for the rational development of SA materials and supply crucial insight into the mechanism of ethylene production through the DCE dechlorination reaction.

The remediation of heavy metal contaminated sediments has become a global consideration, especially the development of economically and environmentally efficient amendments. As a cost-effective natural mineral with nano structure, halloysite nanotubes were coated with carboxylic groups (HNTs@CRC) by a facile and green method, and the fabricated HNTs@CRC was further employed for the stabilization of Co, Ni, Cu and Zn in Maozhou River sediments. After remediation, the results showed an obvious increase in the residual fraction (F4) of the four heavy metals. Moreover, by mixing HNTs@CRC with Ca(OH)2, the metal stabilization performance was further enhanced. In comparison with the raw sediment (no amendment), the F4 percentages were increased by 8.0, 1.4, 2.4 and 1.8 times for Co, Ni, Cu and Zn, respectively. After conducting the toxicity characteristic leaching procedure (TCLP), the stabilization ratio (SR%) was calculated with a value of 98.82% for Co, 83.58% for Ni, 90.71% for Cu and 73.99% for Zn. Additionally, the metal stabilization mechanisms were discussed based on the features of HNTs@CRC and HNTs@CRC/Ca(OH)2, including adsorption, complexation, electrostatic affinity and precipitation. Therefore, by facile and green modification of natural minerals, this study proposed the favorable application of HNTs@CRC, especially HNTs@CRC/Ca(OH)2for the remediation of heavy metal contaminated river sediments.

As a natural clay mineral with fascinating nanostructures, halloysite nanotubes (HNTs) have been widely used as raw materials for the fabrication of HNT-based adsorbents for environmental remediation. However, the current application of the HNT-based adsorbents always suffers from drawbacks such as the use of toxic chemicals during fabrication, relatively low adsorption capacity, and the limitation in the types of adsorbed pollutants. Therefore, in this study, a multifunctional HNT@carbon with rich carboxyl groups (HNT@CRC) was prepared using a facile and cost effective method, and the adsorbent was further used for the removal of pollutants with different features like Pb(ii), Cr(vi) and methylene blue (MB). Results show that the increased amount of carboxyl groups can enhance the adsorption performance of the modified HNT@CRC, and the HNT@CRC-8 exhibited superior and multifunctional adsorption ability for Pb(ii), Cr(vi) and MB with a maximum adsorption capacity of 184.95 mg g−1at pH 6.0, 140.87 mg g−1at pH 2.0 and 690.45 mg g−1at pH 5.5, respectively. By a combination of various techniques (e.g.: X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy) and an extensive comparative analysis, the carboxyl group on the surface of HNT@CRC-8 has been identified to play the most critical role in the efficient removal of Pb(ii), Cr(vi) and MB. Through this work, a new and cost effective method was proposed for the fabrication of a multifunctional HNT@CRC adsorbent with excellent performance for the efficient removal of a variety of pollutants with different features.

Room-temperature high H2sensing has been realized from SnO2/ZnO nanoarray nanogenerator. Without any external electricity power source, the portable device can be self-powered under the driving of human motion, in which the piezoelectric output can actively act as both the power source and H2sensing signal. Upon exposure to 800ppm H2at room temperature, the piezoelectric output voltage of the device under the same applied deformation decreases from 0.80V (in dry air) to 0.14V, and the sensitivity is up to 471.4. The detection limit is ~10ppm H2, and the selectivity against H2at room temperature is very high. The excellent room-temperature H2sensing performance can be attributed to the coupling of the piezoelectric screening effect of ZnO nanowires and the conversion of SnO2/ZnO heterojunctions. This study can stimulate a research trend for the development of the next generation of portable room-temperature H2sensors.

Photocatalytic hydrogen production coupled with selective oxidation of organic substances is of significance and remains a grand challenge because of unsatisfied photocatalytic activity and unclear mechanism. Herein, we first design and synthesize atomically dispersed Pd on ZnIn2S4(PdSA-ZIS) for simultaneously photocatalytic hydrogen production and oxidation of benzylamine. Based on systematic characterization with a combination of theoretical simulations, light irradiation-kelvin probe force microscopy (KPFM) and time-resolved photoluminescence spectrum etc, we demonstrate that Pd single atoms can modulate charge polarization as well as the active sites of ZnIn2S4to promote spatial charge separation and decrease the energy barriers of reaction. The electrons are enriched around Pd single atoms-ZnS layer with lowest energy barriers of H2production, while more holes are accumulated on InS2layer for dehydrogenation of benzylamine. This enables PdSA-ZIS to exhibit remarkable photocatalytic performance for the co-production of N-benzylidenebenzylamine (10.2 mmol g−1h−1) with selectivity of 100% and H2(11.1 mmol g−1h−1) under visible irradiation. Furthermore, by means of controlled experiments, theoretical simulations and in situ electron spin resonance (ESR) spectroscopy, a synergetic photocatalytic redox mechanism on the Pd single atom-ZIS interface for dehydrogenation and coupling of benzylamine and H2production has been investigated in detail.

Lithium-ion capacitors (LICs) bridge the gap between lithium-ion batteries (LIBs) and electrical double-layer capacitors (EDLCs) owing to their unique energy storage mechanisms. From the viewpoints of electrode materials and cell design, the pre-lithiation process is indispensable for improving the working voltage and energy density of LICs. However, the conventional physical short-circuit (PSC) method is time-consuming, which limits the mass-production of practical large-capacity LIC cells. Three alternative pre-lithiation protocols have been proposed, combining the PSC protocol and electrochemical approaches to shorten the pre-lithiation time. The prototype LIC pre-lithiated by using the open-circuit potential cycling (OPC) protocol has the lowest internal resistance and superior high-rate capability (even at 200C-rate). The 900-F large-capacity laminated LIC cells have been assembled and pre-lithiated to validate the feasibility of this method. The pre-lithiation time has been reduced from 470 h (PSC protocol) to 19 h (OPC protocol). This combined protocol is presumed to counteract the voltage loss and enhance the Li+ion diffusion between multiple anode electrodes during the pre-lithiation process.

Photocatalytic selective oxidation of benzylamine into high-value-added organic intermediates holds great promise. Herein, to achieve the goal of efficient separation of photo-generated electrons/holes and regulation of abundant active sites, we systematically investigated the synergistic impact of piezotronic effect and oxygen vacancies on selectively aerobic oxidation of benzylamine into N-benzylidenebenzylamine with Bi4NbO8Br single-crystalline nanosheets (BNO). We found that piezotronic effect and oxygen vacancies could collaboratively enhance the separation of photogenerated electrons and holes. Furthermore, theoretical calculations and experimental results demonstrated that oxygen vacancies can also optimize the adsorption and activation of O2and benzylamine molecules to favor the efficient conversion of which. As a result, the conversion rates of benzylamine increased by about 2.1 times owing to the presence of piezotronic effect and oxygen vacancies. The present work opens an avenue for enhancing photocatalytic activity in selective oxidation of organic pollutant into value-added products, and also shed light on an understanding for the design of a highly efficient multi-field catalytic system in the future.

Photocatalytic hydrogen generation represents a potential approach to address global energy and environmental issues for artificial photosynthesis. Generally, the inevitable use of the holes scavengers leads to increased cost and production of waste. The integration of H2-producing half-reaction with value-added organic molecule oxidation reaction provides a promising strategy to cope with this issue. However, owing to the low separation efficiency of photo-generated charge carriers and sluggish kinetics of the surface reaction, the photocatalytic activity without a co-catalyst is unsatisfying. Herein, the hierarchical and piezoelectric Z-scheme BaTiO3@ZnIn2S4heterostructure without any co-catalyst were dexterously constructed. The Z-scheme electron transfer paths maintain the strong redox ability of the photo-generated electrons and holes, also offer spatial separation of both charge carriers and the surface redox regions. The well-designed redox regions endow BaTiO3@ZnIn2S4with the lowest energy barriers of hydrogen production and C-N coupling of benzylamine than that of pure BaTiO3or ZnIn2S4.Significantly, piezotronic effect can further accelerate the separation and transfer of photo-generated charge carriers in Z-scheme BaTiO3@ZnIn2S4heterostructure. Thus, by the feat of the multiple advantages of the piezotronic effect and Z-scheme heterostructure, the high photocatalytic activity for the co-production of C-N coupling products (5593 umol.g-1) and H2(8041 umol.g-1) was achieved through coupling utilization of mechanical energy and solar light.

The interfacial interaction of activated carbon with volatile organic compounds (VOCs) is seriously affected by water vapor. Therefore, it is vital to enhance the hydrophobic performance of activated carbon for expanding its application in industrial and environmental fields. Herein, a series of hydrophobic activated carbon was fabricated by tailored mixed siloxane and applied in dynamic competitive adsorption at 0, 50, and 90% humidity. Simultaneously, the diffusion molecular models and multicomponent adsorption experiments were used to study the adsorption and diffusion mechanisms. The hydrophobicity of activated carbon was significantly improved by loading of mixed siloxane, in which the equilibrium water absorption decreased from 21.9 to 7.2% and the contact angles increased by 70.10°. Meanwhile, dynamic competitive adsorption at different humidities indicated that the siloxane-functionalized activated carbons (SACs) showed much better competitive adsorption performances for VOCs than original activated carbon, which was further confirmed by the theoretical calculations of adsorption energy. In addition, a remarkable adsorption selectivity and reusability could be demonstrated to VOCs with different polarities on SACs. This study not only provides a new strategy for the hydrophobic modification of activated carbon materials but also offers theoretical guidance for the treatment of gas streams with significant water contents.