Zhejiang University of Science and Technology

In 2008 we published the first set of guidelines for standardizing research in autophagy. Since then, research on this topic has continued to accelerate, and many new scientists have entered the field. Our knowledge base and relevant new technologies have also been expanding. Accordingly, it is important to update these guidelines for monitoring autophagy in different organisms. Various reviews have described the range of assays that have been used for this purpose. Nevertheless, there continues to be confusion regarding acceptable methods to measure autophagy, especially in multicellular eukaryotes. For example, a key point that needs to be emphasized is thatthere is a difference between measurements that monitor the numbers or volume of autophagic elements (e.g., autophagosomes or autolysosomes) at any stage of the autophagic process versus those that measure flux through the autophagy pathway (i.e., the completeprocess including the amount and rate of cargo sequestered and degraded). In particular, a block in macroautophagy that results in autophagosome accumulation must be differentiated from stimuli that increase autophagic activity, defined as increasedautophagy induction coupled with increased delivery to, and degradation within, lysosomes (inmost higher eukaryotes and some protists such as Dictyostelium) or the vacuole (in plants and fungi). In other words, it is especially important that investigators new to the field understand that the appearance of more autophagosomes does not necessarily equate with more autophagy. In fact, in manycases, autophagosomes accumulate because of a block in trafficking to lysosomes without a concomitant change in autophagosome biogenesis, whereas an increase in autolysosomes may reflect a reduction in degradative activity. It is worth emphasizing here that lysosomal digestion is a stage of autophagy and evaluating its competence is a crucial part of the evaluation of autophagic flux, or complete autophagy. Here, we present a set of guidelines for the selection and interpretation of methods for use by investigators who aim to examine macroautophagy and related processes, as well as forreviewers who need to provide realistic and reasonable critiques of papers that are focused on these processes. These guidelines are not meant to be a formulaic set of rules, because the appropriate assays depend in part on the question being asked and the system being used. In addition, we emphasize that no individual assay is guaranteed to be the most appropriate one in every situation, and we strongly recommend the use of multipleassays to monitor autophagy. Along these lines, because of the potential for pleiotropic effects due to blocking autophagy through genetic manipulation, it is imperative to target by gene knockout or RNA interference more than one autophagyrelated protein. In addition, some individual Atg proteins, or groups of proteins, are involved in other cellular pathways implying that not all Atg proteins can be used as a specific marker for an autophagic process. In these guidelines, we consider these various methods of assessing autophagy and what information can, or cannot, be obtained from them. Finally, by discussing the merits and limits of particular assays, we hope to encourage technical innovation in the field.

Recently, a growing interest has been seen in deep learning-based semantic segmentation. UNet, which is one of deep learning networks with an encoder-decoder architecture, is widely used in medical image segmentation. Combining multi-scale features is one of important factors for accurate segmentation. UNet++ was developed as a modified Unet by designing an architecture with nested and dense skip connections. However, it does not explore sufficient information from full scales and there is still a large room for improvement. In this paper, we propose a novel UNet 3+, which takes advantage of full-scale skip connections and deep supervisions. The full-scale skip connections incorporate low-level details with high-level semantics from feature maps in different scales; while the deep supervision learns hierarchical representations from the full-scale aggregated feature maps. The proposed method is especially benefiting for organs that appear at varying scales. In addition to accuracy improvements, the proposed UNet 3+ can reduce the network parameters to improve the computation efficiency. We further propose a hybrid loss function and devise a classification-guided module to enhance the organ boundary and reduce the over-segmentation in a non-organ image, yielding more accurate segmentation results. The effectiveness of the proposed method is demonstrated on two datasets. The code is available at: github.com/ZJUGiveLab/UNet-Version.

Along with the rapid developments in communication technologies and the surge in the use of mobile devices, a brand-new computation paradigm, edge computing, is surging in popularity. Meanwhile, the artificial intelligence (AI) applications are thriving with the breakthroughs in deep learning and the many improvements in hardware architectures. Billions of data bytes, generated at the network edge, put massive demands on data processing and structural optimization. Thus, there exists a strong demand to integrate edge computing and AI, which gives birth to edge intelligence. In this article, we divide edge intelligence into AI for edge (intelligence-enabled edge computing) and AI on edge (artificial intelligence on edge). The former focuses on providing more optimal solutions to key problems in edge computing with the help of popular and effective AI technologies while the latter studies how to carry out the entire process of building AI models, i.e., model training and inference, on the edge. This article provides insights into this new interdisciplinary field from a broader perspective. It discusses the core concepts and the research roadmap, which should provide the necessary background for potential future research initiatives in edge intelligence.

Image dehazing is a representative low-level vision task that estimates latent haze-free images from hazy images. In recent years, convolutional neural network-based methods have dominated image dehazing. However, vision Transformers, which has recently made a breakthrough in high-level vision tasks, has not brought new dimensions to image dehazing. We start with the popular Swin Transformer and find that several of its key designs are unsuitable for image dehazing. To this end, we propose DehazeFormer, which consists of various improvements, such as the modified normalization layer, activation function, and spatial information aggregation scheme. We train multiple variants of DehazeFormer on various datasets to demonstrate its effectiveness. Specifically, on the most frequently used SOTS indoor set, our small model outperforms FFA-Net with only 25% #Param and 5% computational cost. To the best of our knowledge, our large model is the first method with the PSNR over 40 dB on the SOTS indoor set, dramatically outperforming the previous state-of-the-art methods. We also collect a large-scale realistic remote sensing dehazing dataset for evaluating the method's capability to remove highly non-homogeneous haze. We share our code and dataset at https://github.com/IDKiro/DehazeFormer.

Abstract N 6 -methyladenosine (m 6 A) is enriched in 3′untranslated region (3′UTR) and near stop codon of mature polyadenylated mRNAs in mammalian systems and has regulatory roles in eukaryotic mRNA transcriptome switch. Significantly, the mechanism for this modification preference remains unknown, however. Herein we report a characterization of the full m 6 A methyltransferase complex in HeLa cells identifying METTL3/METTL14/WTAP/VIRMA/HAKAI/ZC3H13 as the key components, and we show that VIRMA mediates preferential mRNA methylation in 3′UTR and near stop codon. Biochemical studies reveal that VIRMA recruits the catalytic core components METTL3/METTL14/WTAP to guide region-selective methylations. Around 60% of VIRMA mRNA immunoprecipitation targets manifest strong m 6 A enrichment in 3′UTR. Depletions of VIRMA and METTL3 induce 3′UTR lengthening of several hundred mRNAs with over 50% targets in common. VIRMA associates with polyadenylation cleavage factors CPSF5 and CPSF6 in an RNA-dependent manner. Depletion of CPSF5 leads to significant shortening of 3′UTR of over 2800 mRNAs, 84% of which are modified with m 6 A and have increased m 6 A peak density in 3′UTR and near stop codon after CPSF5 knockdown. Together, our studies provide insights into m 6 A deposition specificity in 3′UTR and its correlation with alternative polyadenylation.

Oxidative stress is associated with many acute and chronic inflammatory diseases, yet limited treatment is currently available clinically. The development of enzyme-mimicking nanomaterials (nanozymes) with good reactive oxygen species (ROS) scavenging ability and biocompatibility is a promising way for the treatment of ROS-related inflammation. Herein we report a simple and efficient one-step development of ultrasmall Cu5.4O nanoparticles (Cu5.4O USNPs) with multiple enzyme-mimicking and broad-spectrum ROS scavenging ability for the treatment of ROS-related diseases. Cu5.4O USNPs simultaneously possessing catalase-, superoxide dismutase-, and glutathione peroxidase-mimicking enzyme properties exhibit cytoprotective effects against ROS-mediated damage at extremely low dosage and significantly improve treatment outcomes in acute kidney injury, acute liver injury and wound healing. Meanwhile, the ultrasmall size of Cu5.4O USNPs enables rapid renal clearance of the nanomaterial, guaranteeing the biocompatibility. The protective effect and good biocompatibility of Cu5.4O USNPs will facilitate clinical treatment of ROS-related diseases and enable the development of next-generation nanozymes. Oxidative stress is involved in several diseases and is a target for intervention. Here, the authors report on the synthesis of ultrasmall copper-based nanozymes as reactive oxygen species scavengers and demonstrate improved treatment outcomes in acute liver and kidney injury and wound healing in vivo.

Water chemistry can be a major factor regulating the toxicity mechanism of ZnO nanoparticles (nano-ZnO) in water. The effect of five commonly used aqueous media with various chemical properties on the toxicity of nano-ZnO to Escherichia coli O111 (E. coli) was investigated, including ultrapure water, 0.85% NaCl, phosphate-buffered saline (PBS), minimal Davis (MD), and Luria-Bertani (LB). Combined results of physicochemical characterization and antibacterial tests of nano-ZnO in the five media suggest that the toxicity of nano-ZnO is mainly due to the free zinc ions and labile zinc complexes. The toxicity of nano-ZnO in the five media deceased as follows: ultrapure water > NaCl > MD > LB > PBS. The generation of precipitates (Zn(3)(PO(4))(2) in PBS) and zinc complexes (of zinc with citrate and amino acids in MD and LB, respectively) dramatically decreased the concentration of Zn(2+) ions, resulting in the lower toxicity in these media. Additionally, the isotonic and rich nutrient conditions improved the tolerance of E. coli to toxicants. Considering the dramatic difference of the toxicity of nano-ZnO in various aqueous media, the effect of water chemistry on the physicochemical properties of nanoparticles should be paid more attention in future nanotoxicity evaluations.

The adsorption of naphthalene, phenanthrene, and pyrene onto graphene (GNS) and graphene oxide (GO) nanosheets was investigated to probe the potential adsorptive sites and molecular mechanisms. The microstructure and morphology of GNS and GO were characterized by elemental analysis, XPS, FTIR, Raman, SEM, and TEM. Graphene displayed high affinity to the polycyclic aromatic hydrocarbons (PAHs), whereas GO adsorption was significantly reduced after oxygen-containing groups were attached to GNS surfaces. An unexpected peak was found in the curve of adsorption coefficients (Kd) with the PAH equilibrium concentrations. The hydrophobic properties and molecular sizes of the PAHs affected the adsorption of G and GO. The high affinities of the PAHs to GNS are dominated by π-π interactions to the flat surface and the sieving effect of the powerful groove regions formed by wrinkles on GNS surfaces. In contrast, the adsorptive sites of GO changed to the carboxyl groups attaching to the edges of GO because the groove regions disappeared and the polar nanosheet surfaces limited the π-π interactions. The TEM and SEM images initially revealed that after loading with PAH, the conformation and aggregation of GNS and GO nanosheets dramatically changed, which explained the observations that the potential adsorption sites of GNS and GO were unusually altered during the adsorption process.

The seminal work of Gatys et al. demonstrated the power of Convolutional Neural Networks (CNNs) in creating artistic imagery by separating and recombining image content and style. This process of using CNNs to render a content image in different styles is referred to as Neural Style Transfer (NST). Since then, NST has become a trending topic both in academic literature and industrial applications. It is receiving increasing attention and a variety of approaches are proposed to either improve or extend the original NST algorithm. In this paper, we aim to provide a comprehensive overview of the current progress towards NST. We first propose a taxonomy of current algorithms in the field of NST. Then, we present several evaluation methods and compare different NST algorithms both qualitatively and quantitatively. The review concludes with a discussion of various applications of NST and open problems for future research. A list of papers discussed in this review, corresponding codes, pre-trained models and more comparison results are publicly available at: https://osf.io/f8tu4/.

In the modern customer‐centred era, customer value is a strategic weapon in attracting and retaining customers. Delivering superior customer value has become a matter of ongoing concern in building and sustaining competitive advantage by driving customer‐relationship‐management (CRM) performance. However, related studies are rather divergent, the key dimensions of customer value remain unclear, and there is no agreement on the evaluation of CRM performance. This paper develops an integrative framework for customer value and CRM performance based on the identification of the key dimensions of customer value. Emphasising the customer equity‐based view, the paper explores the decomposed effects of customer value on CRM performance in terms of relationship quality and customer behaviours. In doing so, a structural equation model is developed using the partial least square method supported by an empirical investigation of customers in China.

With the rapid developments of artificial intelligence (AI) and deep learning (DL) techniques, it is critical to ensure the security and robustness of the deployed algorithms. Recently, the security vulnerability of DL algorithms to adversarial samples has been widely recognized. The fabricated samples can lead to various misbehaviors of the DL models while being perceived as benign by humans. Successful implementations of adversarial attacks in real physical-world scenarios further demonstrate their practicality. Hence, adversarial attack and defense techniques have attracted increasing attention from both machine learning and security communities and have become a hot research topic in recent years. In this paper, we first introduce the theoretical foundations, algorithms, and applications of adversarial attack techniques. We then describe a few research efforts on the defense techniques, which cover the broad frontier in the field. Several open problems and challenges are subsequently discussed, which we hope will provoke further research efforts in this critical area.

Graph neural networks (GNN) has been considered as an attractive modelling method for molecular property prediction, and numerous studies have shown that GNN could yield more promising results than traditional descriptor-based methods. In this study, based on 11 public datasets covering various property endpoints, the predictive capacity and computational efficiency of the prediction models developed by eight machine learning (ML) algorithms, including four descriptor-based models (SVM, XGBoost, RF and DNN) and four graph-based models (GCN, GAT, MPNN and Attentive FP), were extensively tested and compared. The results demonstrate that on average the descriptor-based models outperform the graph-based models in terms of prediction accuracy and computational efficiency. SVM generally achieves the best predictions for the regression tasks. Both RF and XGBoost can achieve reliable predictions for the classification tasks, and some of the graph-based models, such as Attentive FP and GCN, can yield outstanding performance for a fraction of larger or multi-task datasets. In terms of computational cost, XGBoost and RF are the two most efficient algorithms and only need a few seconds to train a model even for a large dataset. The model interpretations by the SHAP method can effectively explore the established domain knowledge for the descriptor-based models. Finally, we explored use of these models for virtual screening (VS) towards HIV and demonstrated that different ML algorithms offer diverse VS profiles. All in all, we believe that the off-the-shelf descriptor-based models still can be directly employed to accurately predict various chemical endpoints with excellent computability and interpretability.

RaptorX Property (http://raptorx2.uchicago.edu/StructurePropertyPred/predict/) is a web server predicting structure property of a protein sequence without using any templates. It outperforms other servers, especially for proteins without close homologs in PDB or with very sparse sequence profile (i.e. carries little evolutionary information). This server employs a powerful in-house deep learning model DeepCNF (Deep Convolutional Neural Fields) to predict secondary structure (SS), solvent accessibility (ACC) and disorder regions (DISO). DeepCNF not only models complex sequence-structure relationship by a deep hierarchical architecture, but also interdependency between adjacent property labels. Our experimental results show that, tested on CASP10, CASP11 and the other benchmarks, this server can obtain ∼84% Q3 accuracy for 3-state SS, ∼72% Q8 accuracy for 8-state SS, ∼66% Q3 accuracy for 3-state solvent accessibility, and ∼0.89 area under the ROC curve (AUC) for disorder prediction.

Generating an image from a given text description has two goals: visual realism and semantic consistency. Although significant progress has been made in generating high-quality and visually realistic images using generative adversarial networks, guaranteeing semantic consistency between the text description and visual content remains very challenging. In this paper, we address this problem by proposing a novel global-local attentive and semantic-preserving text-to-image-to-text framework called MirrorGAN. MirrorGAN exploits the idea of learning text-to-image generation by redescription and consists of three modules: a semantic text embedding module (STEM), a global-local collaborative attentive module for cascaded image generation (GLAM), and a semantic text regeneration and alignment module (STREAM). STEM generates word- and sentence-level embeddings. GLAM has a cascaded architecture for generating target images from coarse to fine scales, leveraging both local word attention and global sentence attention to progressively enhance the diversity and semantic consistency of the generated images. STREAM seeks to regenerate the text description from the generated image, which semantically aligns with the given text description. Thorough experiments on two public benchmark datasets demonstrate the superiority of MirrorGAN over other representative state-of-the-art methods.

Just as graphene triggered a new gold rush, three-dimensional graphene-based macrostructures (3D GBM) have been recognized as one of the most promising strategies for bottom-up nanotechnology and become one of the most active research fields during the last four years. In general, the basic structural features of 3D GBM, including its large surface area, which enhances the opportunity to contact pollutants, and its well-defined porous structure, which facilitates the diffusion of pollutant molecules into the 3D structure, enable 3D GBM to be an ideal material for pollutant management due to its excellent capabilities and easy recyclability. This review aims to describe the environmental applications and mechanisms of 3D GBM and provide perspective. Thus, the excellent performance of 3D GBM in environmental pollutant adsorption, transformation and detection are reviewed. Based on the structures and properties of 3D GBM, the removal mechanisms for dyes, oils, organic solvents, heavy metals, and gas pollutants are highlighted. We attempt to establish "structure-property-application" relationships for environmental pollution management using 3D GBM. Approaches involving tunable synthesis and decoration to regulate the micro-, meso-, and macro-structure and the active sites are also reviewed. The high selectivity, fast rate, convenient management, device applications and recycling utilization of 3D GBM are also emphasized.

As the most frequent wound complication, infection has become a major clinical challenge in wound management. To overcome the "Black Box" status of the wound-healing process, next-generation wound dressings with the abilities of real-time monitoring, diagnosis during early stages, and on-demand therapy has attracted considerable attention. Here, by combining the emerging development of bioelectronics, a smart flexible electronics-integrated wound dressing with a double-layer structure, the upper layer of which is polydimethylsiloxane-encapsulated flexible electronics integrated with a temperature sensor and ultraviolet (UV) light-emitting diodes, and the lower layer of which is a UV-responsive antibacterial hydrogel, is designed. This dressing is expected to provide early infection diagnosis via real-time wound-temperature monitoring by the integrated sensor and on-demand infection treatment by the release of antibiotics from the hydrogel by in situ UV irradiation. The integrated system possesses good flexibility, excellent compatibility, and high monitoring sensitivity and durability. Animal experiment results demonstrate that the integrated system is capable of monitoring wound status in real time, detecting bacterial infection and providing effective treatment on the basis of need. This proof-of-concept research holds great promise in developing new strategies to significantly improve wound management and other pathological diagnoses and treatments.

A novel design method of low profile cavity backed planar slot antenna has been described in this paper. The whole antenna including backed cavity and feeding element is completely constructed at a single substrate by using substrate integrated waveguide technique and grounded coplanar waveguide. An example with 1.7% bandwidth has been presented, which has 5.4 dBi gain, 16.1 dB front-to-back ratio and -19 dB maximum cross polarized radiation level with its total thickness less than lambda <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> /50. The proposed antenna keeps good radiation performance of conventional cavity backed antenna and has advantages of conventional planar antenna including low profile, light weight, easy fabrication with low cost and convenient integration with planar circuit.

Porous organic cages (POCs) are a relatively new class of low-density crystalline materials that have emerged as a versatile platform for investigating molecular recognition, gas storage and separation, and proton conduction, with potential applications in the fields of porous liquids, highly permeable membranes, heterogeneous catalysis, and microreactors. In common with highly extended porous structures, such as metal-organic frameworks (MOFs), covalent organic frameworks (COFs), and porous organic polymers (POPs), POCs possess all of the advantages of highly specific surface areas, porosities, open pore channels, and tunable structures. In addition, they have discrete molecular structures and exhibit good to excellent solubilities in common solvents, enabling their solution dispersibility and processability─properties that are not readily available in the case of the well-established, insoluble, extended porous frameworks. Here, we present a critical review summarizing in detail recent progress and breakthroughs─especially during the past five years─of all the POCs while taking a close look at their strategic design, precise synthesis, including both irreversible bond-forming chemistry and dynamic covalent chemistry, advanced characterization, and diverse applications. We highlight representative POC examples in an attempt to gain some understanding of their structure-function relationships. We also discuss future challenges and opportunities in the design, synthesis, characterization, and application of POCs. We anticipate that this review will be useful to researchers working in this field when it comes to designing and developing new POCs with desired functions.

Abstract Increasing the utilization efficiency of sulfur electrodes and suppressing the “shuttle effect” of intermediate polysulfides are the key challenge for high‐performance lithium–sulfur batteries (LSBs). Herein a facile combined strategy is reported to fabricate novel porous carbon fibers/vanadium nitride arrays (PCF/VN) composite scaffold for the storage of sulfur via a facile chemical etching united solvothermal–supercritical fluid method. More active sulfur can be stored in the PCF/VN backbone and dual blocking effects associated with “physical block and chemical absorption” for polysulfides are achieved in the PCF/VN/S integrated electrode. The PCF with highly porous structure provides large space to accommodate active sulfur and possesses cross‐linked maze channels to physically immobilize the polysulfide species. The VN nanobelt arrays demonstrate strong ability for chemically anchoring the polysulfides, thus retarding the shuttle effect. Due to the unique structure and dual confining effect, the designed PCF/VN/S electrode shows a high reversible capacity of 1310.8 mA h g −1 at 0.1 C, an extended cycle life (1052.5 mA h g −1 after 250 cycles) as well as enhanced rate capability, much better than other counterparts (CF/VN/S, PCF/S, and CF/S). This work opens a new door for fabricating high‐performance integrated electrodes for LSBs.

Abstract Optical logic operations lie at the heart of optical computing, and they enable many applications such as ultrahigh-speed information processing. However, the reported optical logic gates rely heavily on the precise control of input light signals, including their phase difference, polarization, and intensity and the size of the incident beams. Due to the complexity and difficulty in these precise controls, the two output optical logic states may suffer from an inherent instability and a low contrast ratio of intensity. Moreover, the miniaturization of optical logic gates becomes difficult if the extra bulky apparatus for these controls is considered. As such, it is desirable to get rid of these complicated controls and to achieve full logic functionality in a compact photonic system. Such a goal remains challenging. Here, we introduce a simple yet universal design strategy, capable of using plane waves as the incident signal, to perform optical logic operations via a diffractive neural network. Physically, the incident plane wave is first spatially encoded by a specific logic operation at the input layer and further decoded through the hidden layers, namely, a compound Huygens’ metasurface. That is, the judiciously designed metasurface scatters the encoded light into one of two small designated areas at the output layer, which provides the information of output logic states. Importantly, after training of the diffractive neural network, all seven basic types of optical logic operations can be realized by the same metasurface. As a conceptual illustration, three logic operations (NOT, OR, and AND) are experimentally demonstrated at microwave frequencies.