PlantCircRNA: a comprehensive database for plant circular RNAs.

Circular RNAs (circRNAs) represent recently discovered novel regulatory non-coding RNAs. While they are present in many eukaryotes, there has been limited research on plant circRNAs. We developed PlantCircRNA (https://plant.deepbiology.cn/PlantCircRNA/) to fill this gap. The two most important features of PlantCircRNA are (i) it incorporates circRNAs from 94 plant species based on 39 245 RNA-sequencing samples and (ii) it imports the original AtCircDB and CropCircDB databases. We manually curated all circRNAs from published articles, and imported them into the database. Furthermore, we added detailed information of tissue as well as abiotic stresses to the database. To help users understand these circRNAs, the database includes a detection score to measure their consistency and a naming system following the guidelines recently proposed for eukaryotes. Finally, we developed a comprehensive platform for users to visualize, analyze, and download data regarding specific circRNAs. This resource will serve as a home for plant circRNAs and provide the community with unprecedented insights into these mysterious molecule.

Hyaluronic acid-modified mesoporous silica nanoprobes for target identification of atherosclerosis.

Rupture of vulnerable plaque and secondary thrombosis caused by atherosclerosis are one of the main causes of acute cardiovascular and cerebrovascular events, and it is urgent to develop an in-situ, noninvasive, sensitive and targeted detection method at molecular level. We chose CD44, a specific receptor highly expressed on the surface of macrophages, as the target of the molecular probe, and modified the CD44 ligand HA onto the surface of Gd2O3@MSN, constructing the MRI imaging nanoprobe HA-Gd2O3@MSN for targeted recognition of atherosclerosis. The fundamental properties of HA-Gd2O3@MSN were initially investigated. The CCK-8, hemolysis, hematoxylin-eosin staining tests and blood biochemical assays confirmed that HA-Gd2O3@MSN possessed excellent biocompatibility. Laser confocal microscopy, cellular magnetic resonance imaging, flow cytometry and immunohistochemistry were used to verify that the nanoprobes had good targeting properties. The in vivo targeting performance of the nanoprobes was further validated by employing a rabbit atherosclerosis animal model. In summary, the synthesized HA-Gd2O3@MSN nanoprobes have excellent biocompatibility properties as well as good targeting properties. It could provide a new technical tool for early identification of atherosclerosis.

Text-to-Microstructure Generation Using Generative Deep Learning.

Designing novel materials is greatly dependent on understanding the design principles, physical mechanisms, and modeling methods of material microstructures, requiring experienced designers with expertise and several rounds of trial and error. Although recent advances in deep generative networks have enabled the inverse design of material microstructures, most studies involve property-conditional generation and focus on a specific type of structure, resulting in limited generation diversity and poor human-computer interaction. In this study, a pioneering text-to-microstructure deep generative network (Txt2Microstruct-Net) is proposed that enables the generation of 3D material microstructures directly from text prompts without additional optimization procedures. The Txt2Microstruct-Net model is trained on a large microstructure-caption paired dataset that is extensible using the algorithms provided. Moreover, the model is sufficiently flexible to generate different geometric representations, such as voxels and point clouds. The model's performance is also demonstrated in the inverse design of material microstructures and metamaterials. It has promising potential for interactive microstructure design when associated with large language models and could be a user-friendly tool for material design and discovery.

De-icing performance evolution with increasing hydrophobicity by regulating surface topography.

It is of great significance to grasp the role of surface topography in de-icing, which however remains unclear yet. Herein, four textured surfaces are developed by regulating surface topography while keeping surface chemistry and material constituents same. Specifically, nano-textures are maintained and micro-textures are gradually enlarged. The resultant ice adhesion strength is proportional to a topography parameter, i.e. areal fraction of the micro-textures, owing to the localized bonding strengthening, which is verified by ice detachment simulation using finite element method. Moreover, the decisive topography parameter is demonstrated to be determined by the interfacial strength distribution between ice and test surface. Such parameters vary from paper to paper due to different interfacial strength distributions corresponding to respective situations. Furthermore, since hydrophobic and de-icing performance may rely on different topography parameters, there is no certain relationship between hydrophobicity and de-icing.

The role of surface topography in de-icing is verified to be determined by the interfacial strength distribution between ice and surface experimentally and numerically, unveiling the relationship between hydrophobicity and de-icing.

Regulating the Coordination Geometry and Oxidation State of Single-Atom Fe Sites for Enhanced Oxygen Reduction Electrocatalysis.

FeNC catalysts demonstrate remarkable activity and stability for the oxygen reduction reaction (ORR) in polymer electrolyte membrane fuel cells and Zn-air batteries (ZABs). The local coordination of Fe single atoms in FeNC catalysts strongly impacts ORR activity. Herein, FeNC catalysts containing Fe single atoms sites with FeN3 , FeN4 , and FeN5 coordinations are synthesized by carbonization of Fe-rich polypyrrole precursors. The FeN5 sites possess a higher Fe oxidation state (+2.62) than the FeN3 (+2.23) and FeN4 (+2.47) sites, and higher ORR activity. Density functional theory calculations verify that the FeN5 coordination optimizes the adsorption and desorption of ORR intermediates, dramatically lowering the energy barrier for OH- desorption in the rate-limiting ORR step. A primary ZAB constructed using the FeNC catalyst with FeN5 sites demonstrates state-of-the-art performance (an open circuit potential of 1.629 V, power density of 159 mW cm-2 ). Results confirm an intimate structure-activity relationship between Fe coordination, Fe oxidation state, and ORR activity in FeNC catalysts.

Deep-learning-based inverse design of three-dimensional architected cellular materials with the target porosity and stiffness using voxelized Voronoi lattices.

Architected cellular materials are a class of artificial materials with cellular architecture-dependent properties. Typically, designing cellular architectures paves the way to generate architected cellular materials with specific properties. However, most previous studies have primarily focused on a forward design strategy, wherein a geometry is generated using computer-aided design modeling, and its properties are investigated experimentally or via simulations. In this study, we developed an inverse design framework for a disordered architected cellular material (Voronoi lattices) using deep learning. This inverse design framework is a three-dimensional conditional generative adversarial network (3D-CGAN) trained based on supervised learning using a dataset consisting of voxelized Voronoi lattices and their corresponding relative densities and Young's moduli. A well-trained 3D-CGAN adopts variational sampling to generate multiple distinct Voronoi lattices with the target relative density and Young's modulus. Consequently, the mechanical properties of the 3D-CGAN generated Voronoi lattices are validated through uniaxial compression tests and finite element simulations. The inverse design framework demonstrates potential for use in bone implants, where scaffold implants can be automatically generated with the target relative density and Young's modulus.

Gearbox Compound Fault Diagnosis in Edge-IoT Based on Legendre Multiwavelet Transform and Convolutional Neural Network.

The application of edge computing combined with the Internet of Things (edge-IoT) has been rapidly developed. It is of great significance to develop a lightweight network for gearbox compound fault diagnosis in the edge-IoT context. The goal of this paper is to devise a novel and high-accuracy lightweight neural network based on Legendre multiwavelet transform and multi-channel convolutional neural network (LMWT-MCNN) to fast recognize various compound fault categories of gearbox. The contributions of this paper mainly lie in three aspects: The feature images are designed based on the LMWT frequency domain and they are easily implemented in the MCNN model to effectively avoid noise interference. The proposed lightweight model only consists of three convolutional layers and three pooling layers to further extract the most valuable fault features without any artificial feature extraction. In a fully connected layer, the specific fault type of rotating machinery is identified by the multi-label method. This paper provides a promising technique for rotating machinery fault diagnosis in real applications based on edge-IoT, which can largely reduce labor costs. Finally, the PHM 2009 gearbox and Paderborn University bearing compound fault datasets are used to verify the effectiveness and robustness of the proposed method. The experimental results demonstrate that the proposed lightweight network is able to reliably identify the compound fault categories with the highest accuracy under the strong noise environment compared with the existing methods.

Nonionic Surfactant-Assisted In Situ Generation of Stable Passivation Protective Layer for Highly Stable Aqueous Zn Metal Anodes.

A highly stable interface for aqueous rechargeable Zn batteries is of importance to inhibit the growth of Zn dendrites and suppress the side reactions. In this work, we have developed a stable honeycomb-like ZnO passivation protective layer on the Zn surface, which is in situ generated with the assistance of a nonionic surfactant additive (polyethylene glycol tert-octylphenyl ether, denoted as PEGTE). The ZnO passivation layer can facilitate the uniform distribution of the electric field, guiding the uniform deposition of Zn2+ and inhibit the generation of dendrites. As a result, the symmetric cell using the electrolyte with PEGTE shows an excellent performance at high areal capacity, reflected by stable cycling for over 2400 h at 5 mAh/cm2 and 1300 h at 10 mAh/cm2. The full cell paired with V2O5 demonstrates a long lifespan for more than 600 cycles at a low negative/positive capacity ratio.

PSMC2 is overexpressed in glioma and promotes proliferation and anti-apoptosis of glioma cells.

BACKGROUND: This study aims to investigate the effect of PSMC2 expression on the clinical prognosis of glioma patients and its molecular mechanism. METHODS: TCGA multi-tumor screening and survival analysis were combined to explore the differential expression of PSMC2 in multi-tumor. PSMC2 expression in glioma and normal tissues was detected by Western blot and RT-qPCR. Kaplan-Meier survival curve was used to visualize the effect of PSMC2 expression on the overall survival rate and disease-free survival rate of patients with glioma. The highly expressed cell line U343MG was selected to construct a PSMC2 knockdown model by siRNA transfection, and the effect of PSMC2 knockdown on cell proliferation ability was evaluated by CCK-8 assay. Gene-set enrichment analysis of PSMC2 co-expression genes was carried out to predict the molecular mechanism of their regulation of tumor cell phenotypes, and the analysis results were verified by flow cytometry and Western blot. RESULTS: Through broad-spectrum screening of 31 kinds of tumors, we found that PSMC2 was upregulated in most tumors, but PSMC2 was most significantly overexpressed in gliomas and correlated with poor prognosis in glioma patients. The results of Western blot and qRT-PCR showed that PSMC2 was significantly overexpressed in glioma tissues. Further survival analysis revealed that the overall survival and disease-free survival of patients with low PSMC2 expression were significantly better than that of patients with high PSMC2 expression. The proliferation of U343MG cells was significantly inhibited after PSMC2 knockdown. Enrichment analysis of PSMC2 co-expression genes indicated that PSMC2 affected the apoptosis process. The expression of apoptosis-related proteins also significantly changed following PSMC2 knockdown. CONCLUSIONS: PSMC2 promotes the proliferation of glioma cells and inhibits the apoptosis, which is expected to be a potential therapeutic target for glioma.

Propofol suppresses microglial phagocytosis through the downregulation of MFG-E8.

BACKGROUND: Microglia are highly motile phagocytic cells in the healthy brain with surveillance and clearance functions. Although microglia have been shown to engulf cellular debris following brain insult, less is known about their phagocytic function in the absence of injury. Propofol can inhibit microglial activity, including phagocytosis. Milk fat globule epidermal growth factor 8 (MFG-E8), as a regulator of microglia, plays an essential role in the phagocytic process. However, whether MFG-E8 affects the alteration of phagocytosis by propofol remains unknown. METHODS: Microglial BV2 cells were treated with propofol, with or without MFG-E8. Phagocytosis of latex beads was evaluated by flow cytometry and immunofluorescence. MFG-E8, p-AMPK, AMPK, p-Src, and Src levels were assessed by western blot analysis. Compound C (AMPK inhibitor) and dasatinib (Src inhibitor) were applied to determine the roles of AMPK and Src in microglial phagocytosis under propofol treatment. RESULTS: The phagocytic ability of microglia was significantly decreased after propofol treatment for 4 h (P < 0.05). MFG-E8 production was inhibited by propofol in a concentration- and time-dependent manner (P < 0.05). Preadministration of MFG-E8 dose-dependently (from 10 to 100 ng/ml) reversed the suppression of phagocytosis by propofol (P < 0.05). Furthermore, the decline in p-AMPK and p-Src levels induced by propofol intervention was reversed by MFG-E8 activation (P < 0.05). Administration of compound C (AMPK inhibitor) and dasatinib (Src inhibitor) to microglia blocked the trend of enhanced phagocytosis induced by MFG-E8 (P < 0.05). CONCLUSIONS: These findings reveal the intermediate role of MFG-E8 between propofol and microglial phagocytic activity. Moreover, MFG-E8 may reverse the suppression of phagocytosis induced by propofol through the regulation of the AMPK and Src signaling pathways.

Head and neck CT angiography to assess the internal carotid artery stealing pathway.

BACKGROUND: Common carotid artery occlusive disease (CCAOD) could form internal carotid artery steal pathways. Based on the diagnostic results of digital subtraction angiography (DSA), head and neck computed tomography angiography (CTA) was used to find the internal carotid artery stealing pathway after CCAOD. METHODS: The clinical and imaging data of 18 patients with CCAOD were retrospectively analyzed. DSA and CTA was used to evaluate internal carotid artery steal pathways. RESULTS: Of the 18 patients with CCAOD, 10 patients found internal carotid artery steal pathways. There were 7 males and 3 females. Vascular ultrasound examination of all patients: The affected side had no blood flow in common carotid artery (CCA), and had retrograde blood flow in the external carotid artery (ECA). The blood flow of the affected side was decreased in the internal carotid artery (ICA), but it was antegrade. DSA diagnosed 10 cases of CCA occlusion and CTA diagnosed 10 cases of CCA occlusion. DSA and CTA found 6 internal carotid artery blood stealing pathways: ① Vertebral artery → occipital artery → external carotid artery → internal carotid artery (6 cases); ② Thyrocervical trunk → ascending cervical artery → occipital artery → external carotid artery → internal carotid artery (7 cases); ③ Costocervical trunk → deep cervical artery → occipital artery → external carotid artery → internal carotid artery (6 cases); ④ Affected side thyroid neck trunk → inferior thyroid artery → superior thyroid artery → external carotid artery → internal carotid artery (2 cases); ⑤ Contralateral external carotid artery → contralateral superior thyroid artery → affected superior thyroid artery → external carotid artery → neck Internal artery (2 cases); ⑥ Parathyroid neck → superficial cervical artery → occipital artery → external carotid artery → internal carotid artery (1 case). CONCLUSIONS: The patients with CCAOD can find the internal carotid artery blood stealing pathway through CTA.

Perioperative Dexmedetomidine attenuates brain ischemia reperfusion injury possibly via up-regulation of astrocyte Connexin 43.

BACKGROUND: Astrocyte Connexin 43 (Cx43) is essential for the trophic and protective support of neurons during brain ischemia reperfusion (I/R) injury. It is believed that dexmedetomidine participates in Cx43-mediated effects. However, its mechanisms remained unclear. This study aims to address the relationship and regulation among them. METHODS: Adult male Sprague-Dawley rats were allocated to the 90-min right middle cerebral arterial occlusion with or without dexmedetomidine pretreatment (5 µg/kg). Neurological functions were evaluated and brain lesions, as well as inflammatory factors (IL-1ß, IL-6, TNF-α), were assessed. Ischemic penumbral cortex was harvested to determine the expression of astrocyte Cx43. Primary astrocytes were cultured to evaluate the effect of dexmedetomidine on Cx43 after oxygen-glucose deprivation. RESULTS: Dexmedetomidine pretreatment attenuated neurological injury, brain lesions and expression of inflammatory factors (IL-1ß, IL-6, TNF-α) after brain ischemia (P < 0.05). Astrocyte Cx43 was down-regulated by brain I/R injury, both in vivo and in vitro, which were reversed by dexmedetomidine (P < 0.05). This effect was mediated by the phosphorylation of Akt and GSK-3ß. Further studies with LY294002 (PI3K inhibitor) or SB216763 (GSK-3ß inhibitor) confirmed the effect of dexmedetomidine on astrocyte Cx43. CONCLUSIONS: Perioperative dexmedetomidine administration attenuates neurological injury after brain I/R injury, possibly through up-regulation of astrocyte Cx43. Activation of PI3K-Akt-GSK-3ß pathway might contribute to this protective effect.

Circular RNA circ_0020123 promotes non-small cell lung cancer progression by acting as a ceRNA for miR-488-3p to regulate ADAM9 expression.

Circular RNAs (circRNAs) is a class of non-coding RNA with important functions in tumor development and progression. In the previous study, circ_0020123 was found to be an elevated circRNA in non-small cell lung cancer (NSCLC) tissues screened by high-throughput sequencing. In the current work, we uncovered the clinical significance, biological functions and mechanism of circ_0020123 in NSCLC. The expression profile of circ_0020123 was measured by RT-qPCR. Correlations between circ_0020123 expression and patients' clinical features were evaluated. Cell viability, apoptosis, migration and invasion were detected by cell counting kit-8 (CCK-8), flow cytometric and transwell assay, respectively. Bioinformatics database and luciferase reporter gene assays were utilized to identify the mechanisms of circ_0020123. The results revealed elevation of circ_0020123 in tissue specimens and cells than the nontumorous tissues and 16HBE, separately. The enhancement of circ_0020123 in tumor tissues correlated with positive lymph node metastasis, advanced TNM stages, and adverse prognosis for NSCLC patients. Functionally, silencing of circ_0020123 distinctly suppressed the growth, migration and invasion and inhibited the apoptosis of A549 cells. On the contrary, when circ_0020123 expression was ectopically expressed, the above effects were significantly strengthened in H1299 cell line. For the mechanism exploration, circ_0020123 could sponge miR-488-3p to release its inhibition on ADAM9 expression. Moreover, the functional role of circ_0020123 is partly dependent on its regulation of ADAM9 proved by rescue assays. Taken together, our findings provide the possibility that circ_0020123 may be a new target for NSCLC prognosis prediction and therapy.

Temperature and CO2 Dual-Responsive Pickering Emulsions Using Jeffamine M2005-Modified Cellulose Nanocrystals.

Cellulose nanocrystals (CNCs) with excellent biodegradability are promising biomaterials for use as responsive Pickering emulsifiers. However, the high hydrophilicity of CNCs limits their emulsification ability. Some existing studies have utilized complicated covalent modification procedures to increase the hydrophobicity of CNCs. To simplify the modification process, we prepared hydrophobically modified CNCs (CNCs-M2005) via simple and controllable electrostatic interactions with thermosensitive M2005. The obtained CNCs-M2005 exhibited temperature and CO2 dual-responsive properties. Subsequently, stable oil/water Pickering emulsions were prepared using the partially hydrophobic CNCs-M2005 at 20 °C. However, demulsification occurred when the temperature increased to 60 °C. This temperature-induced demulsification resulted from the dehydration of polyethylene oxide and polypropylene oxide, causing the aggregation of the CNCs-M2005, as shown by dynamic light scattering and transmission electron microscopy experiments. In addition, demulsification was also achieved after bubbling CO2, which was attributed to the dissociation of the partially hydrophobic CNCs-M2005. The temperature and CO2 dual-responsive biosafe Pickering emulsions open up opportunity for the design of intelligent food, cosmetic, and drug delivery systems.

Enhanced production of xylose from corncob hydrolysis with oxalic acid as catalyst.

The acid-catalyzed treatment was a conventional process for xylose production from corncob. To increase the release of xylose and to reduce the by-products formation and water usage, the oxalic acid was used as catalyst to hydrolyze the corncob and the hydrolytic conditions were investigated. The highest xylose yield of 32.7 g L-1, representing 96.1% of total theoretical xylose yield, was obtained using 1.2% oxalic acid after hydrolysis for 120 min at 130 °C, which was more than 10% higher than that of sulfuric acid-catalyzed hydrolysis. Mixed acids-catalyzed hydrolysis performed a synergistic effect for xylose production and 31.7 g L-1 of xylose was reached after reacting for 90 min with oxalic acid and sulfuric acid at a ratio of 1:4 (w/w). A kinetic model was developed to elucidate the competitive reaction between xylose formation and its degradation in the hydrolysis process, and the experimental data obtained in this study were perfectly in agreement with that of predicted from the model. Furthermore, the final xylose yield of 85% was achieved after purification and crystallization. It was demonstrated that xylose production from the corncob hydrolysis with oxalic acid as the catalyst was an effective alternative to the traditional sulfuric acid-based hydrolysis.

Site-specific antibody-drug conjugation through glycoengineering.

Antibody-drug conjugates (ADCs) have been proven clinically to be more effective anti-cancer agents than native antibodies. However, the classical conjugation chemistries to prepare ADCs by targeting primary amines or hinge disulfides have a number of shortcomings including heterogeneous product profiles and linkage instability. We have developed a novel site-specific conjugation method by targeting the native glycosylation site on antibodies as an approach to address these limitations. The native glycans on Asn-297 of antibodies were enzymatically remodeled in vitro using galactosyl and sialyltransferases to introduce terminal sialic acids. Periodate oxidation of these sialic acids yielded aldehyde groups which were subsequently used to conjugate aminooxy functionalized cytotoxic agents via oxime ligation. The process has been successfully demonstrated with three antibodies including trastuzumab and two cytotoxic agents. Hydrophobic interaction chromatography and LC-MS analyses revealed the incorporation of ~1.6 cytotoxic agents per antibody molecule, approximating the number of sialic acid residues. These glyco-conjugated ADCs exhibited target-dependent antiproliferative activity toward antigen-positive tumor cells and significantly greater antitumor efficacy than naked antibody in a Her2-positive tumor xenograft model. These findings suggest that enzymatic remodeling combined with oxime ligation of the native glycans of antibodies offers an attractive approach to generate ADCs with well-defined product profiles. The site-specific conjugation approach presented here provides a viable alternative to other methods, which involve a need to either re-engineer the antibody sequence or develop a highly controlled chemical process to ensure reproducible drug loading.

Multifunctional Separator Enables High-Performance Sodium Metal Batteries in Carbonate-Based Electrolytes.

Sodium metal has become one of the most promising anodes for next-generation cheap and high-energy-density metal batteries; however, challenges caused by the uncontrollable sodium dendrite growth and fragile solid electrolyte interphase (SEI) restrict their large-scale practical applications in low-cost and wide-voltage-window carbonate electrolytes. Herein, a novel multifunctional separator with lightweight and high thinness is proposed, assembled by the cobalt-based metal-organic framework nanowires (Co-NWS), to replace the widely applied thick and heavy glass fiber separator. Benefitting from its abundant sodiophilic functional groups and densely stacked nanowires, Co-NWS not only exhibits outstanding electrolyte wettability and effectively induces uniform Na+ ion flux as a strong ion redistributor but also favors constructing the robust N,F-rich SEI layer. Satisfactorily, with 10 µL carbonate electrolyte, a Na|Co-NWS|Cu half-cell delivers stable cycling (over 260 cycles) with a high average Coulombic efficiency of 98%, and the symmetric cell shows a long cycle life of more than 500 h. Remarkably, the full cell shows a long-term life span (over 1500 cycles with 92% capacity retention) at high current density in the carbonate electrolyte. This work opens up a strategy for developing dendrite-free, low-cost, and long-life-span sodium metal batteries in carbonate-based electrolytes.

Glycan structure determinants for cation-independent mannose 6-phosphate receptor binding and cellular uptake of a recombinant protein.

The cation-independent mannose 6-phosphate receptor (CI-MPR) plays a critical role in intracellular transport of lysosomal enzymes as well as the uptake of recombinant proteins. To define the minimal glycan structure determinants necessary for receptor binding and cellular uptake, we synthesized a series of glycans containing mono-, di-, tri-, tetra-, and hexamannoses terminated with either one or two phosphates for conjugating to a model protein, recombinant human acid α-glucosidase. A high affinity interaction with the CI-MPR can be achieved for the enzyme conjugated to a dimannose glycan with a single phosphate. However, tightest binding to a CI-MPR affinity column was observed with a hexamannose structure containing two phosphates. Moreover, maximal cellular uptake and a 5-fold improvement in in vivo potency were achieved when the bisphosphorylated hexamannose glycan is conjugated to the protein by a ß linker. Nevertheless, even a monophosphorylated dimannose glycan conjugate showed stronger binding to the receptor affinity column, higher cellular uptake, and significantly greater in vivo efficacy compared to the unconjugated protein which contains a low level of high affinity glycan structure. These results demonstrate that the phosphorylated dimannose moiety appears to be the minimal structure determinant for enhanced CI-MPR binding and that the orientation of the glycan is critical for maximum receptor interaction. In summary, we have improved the understanding of the mechanism of CI-MPR binding and developed a simple alternative for CI-MPR targeting.

Deep Learning in Mechanical Metamaterials: From Prediction and Generation to Inverse Design.

Mechanical metamaterials are meticulously designed structures with exceptional mechanical properties determined by their microstructures and constituent materials. Tailoring their material and geometric distribution unlocks the potential to achieve unprecedented bulk properties and functions. However, current mechanical metamaterial design considerably relies on experienced designers' inspiration through trial and error, while investigating their mechanical properties and responses entails time-consuming mechanical testing or computationally expensive simulations. Nevertheless, recent advancements in deep learning have revolutionized the design process of mechanical metamaterials, enabling property prediction and geometry generation without prior knowledge. Furthermore, deep generative models can transform conventional forward design into inverse design. Many recent studies on the implementation of deep learning in mechanical metamaterials are highly specialized, and their pros and cons may not be immediately evident. This critical review provides a comprehensive overview of the capabilities of deep learning in property prediction, geometry generation, and inverse design of mechanical metamaterials. Additionally, this review highlights the potential of leveraging deep learning to create universally applicable datasets, intelligently designed metamaterials, and material intelligence. This article is expected to be valuable not only to researchers working on mechanical metamaterials but also those in the field of materials informatics.

Deep Spiking Residual Shrinkage Network for Bearing Fault Diagnosis.

Bearing fault diagnosis of electrical equipment has been a popular research area in recent years because there are often some faults during continuous operation in production due to the harsh working environment. However, the traditional fault signal processing methods rely on highly expert experience, and some parameters are difficult to be optimized by machine-learning methods. Thus, the satisfactory recognition accuracy of fault diagnosis cannot be achieved in the above methods. In this article, a new model based on the spiking neural network (SNN) is proposed, which is called deep the spiking residual shrinkage network (DSRSN) for bearing fault diagnosis. In the model, attention mechanisms and soft thresholding are introduced to improve the recognition rate under a high-level noise background. The higher recognition accuracy is obtained in the proposed model which is tested on the fault signal dataset under different noise intensities. Meanwhile, the training time is about treble as fast as the training time of the artificial neural network, which is reflecting the high efficiency of SNN.