Cardiac-specific deletion of BRG1 ameliorates ventricular arrhythmia in mice with myocardial infarction.

Malignant ventricular arrhythmia (VA) after myocardial infarction (MI) is mainly caused by myocardial electrophysiological remodeling. Brahma-related gene 1 (BRG1) is an ATPase catalytic subunit that belongs to a family of chromatin remodeling complexes called Switch/Sucrose Non-Fermentable Chromatin (SWI/SNF). BRG1 has been reported as a molecular chaperone, interacting with various transcription factors or proteins to regulate transcription in cardiac diseases. In this study, we investigated the potential role of BRG1 in ion channel remodeling and VA after ischemic infarction. Myocardial infarction (MI) mice were established by ligating the left anterior descending (LAD) coronary artery, and electrocardiogram (ECG) was monitored. Epicardial conduction of MI mouse heart was characterized in Langendorff-perfused hearts using epicardial optical voltage mapping. Patch-clamping analysis was conducted in single ventricular cardiomyocytes isolated from the mice. We showed that BRG1 expression in the border zone was progressively increased in the first week following MI. Cardiac-specific deletion of BRG1 by tail vein injection of AAV9-BRG1-shRNA significantly ameliorated susceptibility to electrical-induced VA and shortened QTc intervals in MI mice. BRG1 knockdown significantly enhanced conduction velocity (CV) and reversed the prolonged action potential duration in MI mouse heart. Moreover, BRG1 knockdown improved the decreased densities of Na+ current (INa) and transient outward potassium current (Ito), as well as the expression of Nav1.5 and Kv4.3 in the border zone of MI mouse hearts and in hypoxia-treated neonatal mouse ventricular cardiomyocytes. We revealed that MI increased the binding among BRG1, T-cell factor 4 (TCF4) and ß-catenin, forming a transcription complex, which suppressed the transcription activity of SCN5A and KCND3, thereby influencing the incidence of VA post-MI.

A Lotus Seedpods-Inspired Interfacial Solar Steam Generator with Outstanding Salt Tolerance and Mechanical Properties for Efficient and Stable Seawater Desalination.

Interfacial solar steam generators (ISSGs) can capture solar energy and concentrate the heat at the gas-liquid interface, resulting in efficient water evaporation. However, traditional ISSGs have limitations in long-term seawater desalination processes, such as limited light absorption area, slow water transport speed, severe surface salt accumulation, and weak mechanical performance. Inspired by lotus seedpods, a novel ISSG (rGO-SA-PSF) is developed by treating a 3D warp-knitted spacer fabric with plasma (PSF) and combining it with sodium alginate (SA) and reduces graphene oxide (rGO). The rGO-SA-PSF utilizes a core-suction effect to achieve rapid water pumping and employs aerogel to encapsulate the plasma-treated spacer yarns to create the lotus seedpod-inspired hydrophilic stems, innovatively constructing multiple directional water transport channels. Simultaneously, the large holes of rGO-SA-PSF on the upper layer form lotus seedpod-inspired head concave holes, enabling efficient light capture. Under 1 kW m-2 illumination, rGO-SA-PSF exhibits a rapid evaporation rate of 1.85 kg m-2  h-1 , with an efficiency of 96.4%. Additionally, it shows superior salt tolerance (with no salt accumulation during continuous evaporation for 10 h in 10% brine) and self-desalination performance during long-term seawater desalination processes. This biomimetic ISSG offers a promising solution for efficient and stable seawater desalination and wastewater purification.

Metabolic engineering of Escherichia coli for the production of an antifouling agent zosteric acid.

Zosteric acid (ZA) is a Zostera species-derived, sulfated phenolic acid compound with antifouling activity and has gained much attention due to its nontoxic and biodegradable characteristics. However, the yield of Zostera species available for ZA extraction is limited by natural factors, such as season, latitude, light, and temperature. Here we report the development of metabolically engineered Escherichia coli strains capable of producing ZA from glucose and glycerol. First, intracellular availability of the sulfur donor 3'-phosphoadenosine-5'-phosphosulfate (PAPS) was enhanced by knocking out the cysH gene responsible for PAPS consumption and overexpressing the genes required for PAPS biosynthesis. Co-overexpression of the genes encoding tyrosine ammonia-lyase, sulfotransferase 1A1, ATP sulfurylase, and adenosine 5'-phosphosulfate kinase constructed ZA producing strain with enhanced PAPS supply. Second, the feedback-resistant forms of aroG and tyrA genes (encoding 3-deoxy-d-arabinoheptulosonate 7-phosphate synthase and chorismate mutase, respectively) were overexpressed to relieve the feedback regulation of L-tyrosine biosynthesis. Third, the pykA gene involved in phosphoenolpyruvate-consuming reaction, the regulator gene tyrR, the competing pathway gene pheA, and the ptsHIcrr genes essential for the PEP:carbohydrate phosphotransferase system were deleted. Moreover, all genes involved in the shikimate pathway and the talA, tktA, and tktB genes in the pentose phosphate pathway were examined for ZA production. The PTS-independent glucose uptake system, the expression vector system, and the carbon source were also optimized. As a result, the best-performing strain successfully produced 1.52 g L-1 ZA and 1.30 g L-1p-hydroxycinnamic acid from glucose and glycerol in a 700 mL fed-batch bioreactor.

Plasma exosome-derived connexin43 as a promising biomarker for melanoma patients.

BACKGROUND: To examine the levels of exosome-derived connexin 43 (Cx43) in plasma and estimate its forecast value in patients with melanoma. METHODS: We measured the plasma exosome-derived Cx43 levels in the plasma of 112 melanoma patients and 50 healthy controls. RESULTS: The plasma exosome-derived Cx43 levels in patients with melanoma were substantially downregulated as opposed to the levels in healthy controls (P < 0.001). Kaplan-Meier analysis indicated that overall survival (OS) and disease-free survival (DFS) were poorer in patients with melanoma who exhibited lower levels of plasma exosome-derived Cx43 (both P < 0.001). The levels of plasma exosome-derived Cx43 were considerably elevated in patients with melanoma whose tumor was situated in the skin, tumor size < 10 cm, with Clark level I-III, TNM stages IIb-IV, and had no lymph node metastasis as opposed to patients whose tumor was situated in the viscera or mucosa, tumor size ≥ 10 cm, Clark level IV-V, TNM stages IIb-IV and had lymph node metastasis (all P < 0.05). The receiver operating characteristic (ROC) of plasma exosome-derived Cx43 for forecasting 5-year DFS in patients with melanoma was 0.78 (95% confidence interval (CI): 0.70-0.86), with a specificity of 77.78% and a sensitivity of 81.55%. The ROC of plasma exosome-derived Cx43 for forecasting 5-year OS of patients with melanoma was 0.77 (95% CI: 0.68-0.84), with a specificity of 80.0% and sensitivity of 65.98%. CONCLUSION: The overall findings indicated that the levels of plasma exosome-derived Cx43 in patients with melanoma were considerably downregulated. It can therefore be inferred that the levels of plasma exosome-derived Cx43 might be a prospective prognostic indicator for 5 5-year OS and 5-year DFS of patients with melanoma.

Structure-based mining of a chitosanase with distinctive degradation mode and product specificity.

Chitosan derivates with varying degrees of polymerization (DP) have attracted great concern due to their excellent biological activities. Increasing the abundance of chitosanases with different degradation modes contributes to revealing their catalytic mechanisms and facilitating the production of chitosan derivates. However, the identification of endo-chitosanases capable of producing chitobiose and D-glucosamine (GlcN) from chitosan substrates has remained elusive. Herein, an endo-chitosanase (CsnCA) belonging to the GH46 family was identified based on structural analysis in phylogenetic evolution. Moreover, we demonstrate that CsnCA acts in a random endo-acting manner, producing chitosan derivatives with DP ≤ 2. The in-depth analysis of CsnCA revealed that (GlcN)3 serves as the minimal substrate, undergoing cleavage in the mode that occupies the subsites - 2 to + 1, resulting in the release of GlcN. This study succeeded in discovering a chitosanase with distinctive degradation modes, which could facilitate the mechanistic understanding of chitosanases, further empowering the production of chitosan derivates with specific DP. KEY POINTS: • Structural docking and evolutionary analysis guide to mining the chitosanase. • The endo-chitosanase exhibits a unique GlcN-producing cleavage pattern. • The cleavage direction of chitosanase to produce GlcN was identified.

Dataglove for Sign Language Recognition of People with Hearing and Speech Impairment via Wearable Inertial Sensors.

Finding ways to enable seamless communication between deaf and able-bodied individuals has been a challenging and pressing issue. This paper proposes a solution to this problem by designing a low-cost data glove that utilizes multiple inertial sensors with the purpose of achieving efficient and accurate sign language recognition. In this study, four machine learning models-decision tree (DT), support vector machine (SVM), K-nearest neighbor method (KNN), and random forest (RF)-were employed to recognize 20 different types of dynamic sign language data used by deaf individuals. Additionally, a proposed attention-based mechanism of long and short-term memory neural networks (Attention-BiLSTM) was utilized in the process. Furthermore, this study verifies the impact of the number and position of data glove nodes on the accuracy of recognizing complex dynamic sign language. Finally, the proposed method is compared with existing state-of-the-art algorithms using nine public datasets. The results indicate that both the Attention-BiLSTM and RF algorithms have the highest performance in recognizing the twenty dynamic sign language gestures, with an accuracy of 98.85% and 97.58%, respectively. This provides evidence for the feasibility of our proposed data glove and recognition methods. This study may serve as a valuable reference for the development of wearable sign language recognition devices and promote easier communication between deaf and able-bodied individuals.

Detection of Listeria monocytogenes in Food Using the Proofman-LMTIA Assay.

Microbial factors, including bacteria, viruses, and other pathogens, are significant contributors to foodborne illnesses, posing serious food safety risks due to their potential for rapid growth and contamination. Listeria monocytogenes is one of the most common types of foodborne bacteria that can cause serious foodborne diseases or even fatalities. In this study, a novel nucleic acid amplification method called Proofman-LMTIA was employed to detect Listeria monocytogenes contamination in food. This method combines proofreading enzyme-mediated probe cleavage with ladder-shape melting temperature isothermal amplification. A positive recombinant plasmid was used as a control to ensure the accuracy of the detection results, and primers and Proofman probes were specifically designed for the LMTIA. Genomic DNA was extracted, the reaction temperature was optimized, and the primers' specificity was verified using foodborne pathogens like Staphylococcus aureus, Escherichia coli O157:H7, and Salmonella. The sensitivity was assessed by testing serial dilutions of genomic DNA, and the method's applicability was confirmed by detecting artificially contaminated fresh pork. The established LMTIA method exhibited both high specificity and sensitivity. At the optimal reaction temperature of 63 °C, the primers specifically identified Listeria monocytogenes contamination in pork at a concentration of 8.0 ± 0.7 colony-forming units (CFUs) per 25 g. Furthermore, the Proofman-LMTIA method was applied to test Listeria monocytogenes DNA in 30 food samples purchased from a Chinese retail market, and reassuringly, all results indicated no contamination. Proofman-LMTIA can serve as a reliable and rapid method for detecting Listeria monocytogenes in food, contributing to public health by safeguarding consumers from foodborne illnesses, and strengthening food safety regulations.

Mixotrophic Cultivation of Microalgae Using Biogas as the Substrate.

Biogas utilization through biotechnology represents a potential and novel technology. We propose the microalgal mixotrophic cultivation to convert biogas to microalgae-based biodiesel, in which methanotroph was co-cultured to convert CH4 to organic intermediate (and CO2) for microalgal mixotrophic growth. This study constructed a co-culture of Methylocystis bryophila (methanotroph) and Scenedesmus obliquus (microalgae) with biogas feeding. Compared with the single culture of S. obliquus, higher microalgal biomass but a lower chlorophyll concentration was observed. The organic metabolism-related genes were upregulated, verifying microalgal mixotrophic growth. The stoichiometric calculation of M. bryophila culture shows that M. bryophila tends to release organic matter rather than grow under a low O2 content. M. bryophila rarely grew under five different light intensities, indicating that M. bryophila acts as a biocatalyst in the co-culture. The organic intermediate released by methanotroph increased the maximum biomass of microalgal culture, accelerated nitrogen absorption, accumulated more monounsaturated fatty acids, and improved the adaptation to light. The co-culture of microalgae and methanotroph may provide new opportunities for microalgae-based biodiesel production using biogas as a substrate.

Development of a Ladder-Shape Melting Temperature Isothermal Amplification (LMTIA) Assay for the Identification of Cassava Component in Sweet Potato Starch Noodles.

Food authenticity has become increasingly important as a result of food adulteration. To identify the authenticity of sweet potato starch noodles, the ladder-shape melting temperature isothermal amplification (LMTIA) method of determining cassava (Manihot esculenta Crantz) DNA in sweet potato starch noodles was used. A set of primers targeted at the internal transcription spacer (ITS) of cassava was designed, genomic DNA was extracted, the LMTIA reaction temperature was optimized, and the specificity of the primer was verified with the genomic DNAs of cassava, sweet potato (Ipomoea batatas L.), Solanum tuberosum L., Zea mays L., Vigna radiate L., Triticum aestivum L., and Glycine max (L.) Merr. The sensitivity with the serially diluted genomic DNA of cassava and the suitability for the DNA extracted from sweet potato starch adulterated with cassava starch were tested. The LMTIA assay for identifying the cassava component in sweet potato starch noodles was established. At the optimal temperature of 52 °C, the primers could specifically distinguish a 0.01% (w/w) cassava component added to sweet potato starch. Additionally, the LMTIA method was applied to the cassava DNA detection of 31 sweet potato starch noodle samples purchased from retail markets in China. Of these, 14 samples were positive. The LMTIA assay could be a reliable method for the rapid detection of cassava components in sweet potato starch noodles, to protect the rights of consumers and to regulate the sale market order of starch noodles.

Carbon quantum Dot@Silver nanocomposite-based fluorescent imaging of intracellular superoxide anion.

Silver nanoparticle (Ag NP)-coated carbon quantum dot (CQD) core-shell-structured nanocomposites (CQD@Ag NCs) were developed for fluorescent imaging of intracellular superoxide anion (O2•-). The morphology of CQD@Ag NCs was investigated by transmission electron microscopy, and the composition was characterized by X-ray diffraction and X-ray photoelectron spectroscopy. CQDs display blue fluorescence with excitation/emission maxima at 360/440 nm, and the fluorescence was quenched by Ag NPs in CQD@Ag NCs. In the presence of O2•-, Ag NPs were oxide-etched and the fluorescence of CQDs was recovered. A linearity between the relative fluorescence intensity and O2•- solution concentration within the range 0.6 to 1.6 µM was found, with a detection limit of 0.3 µM. Due to their high sensitivity, selectivity, and low cytotoxicity, the as-synthesized CQD@Ag NCs have been successfully applied for imaging of O2•- in MCF-7 cells during the whole process of autophagy induced by serum starvation. In our perception, the developed method provides a cost-effective, sensitive, and selective tool in bioimaging and monitoring of intracellular O2•- changes, and is promising for potential biological applications. Graphical abstract Illustration of the synthesis of carbon quantum Dot@Silver nanocomposites (CQD@Ag NCs), and CQD@Ag NCs as a "turn-on" nanoprobe for fluorescent imaging of intracellular superoxide anion.

The Regenerated Tissue at the Donor Site After Costal Cartilage Harvest for Auricular Reconstruction.

OBJECTIVE: To observe and summarize the nature of the regenerative tissue at the donor site after harvesting costal cartilage for auricular reconstruction and to explore the contribution of the perichondrium to the regeneration of costal cartilage in the clinic. METHODS: From January 2016 to June 2017, 23 patients with microtia who were performed chest computed tomography (CT) after costal cartilage harvest for ear reconstruction were reviewed. And they had the surgery for at least 6 months. Of 23 patients, 17 patients were males and 6 were females; these patients were aged 7 to 43 years (mean age, 15.2 years). The authors divided the patients into 2 groups according to whether the perichondrium was retained or not. Group 1 was patients with intact perichondrium, total 20. Group 2 was patients with damaged perichondrium, total 3. Every patients' regenerative tissue CT value at the donor-site region of costal cartilage was measured and recorded. In addition, 2 regenerated tissue samples for examined histologic evaluation by hematoxylin and eosin stain were collected. RESULTS: Of 23 patients, regenerated tissue with high CT value (above 100 Hounsfield unit [Hu]) was observed in 19 (82.61%) patients from group 1. And the direction of the regenerated tissue is roughly similar to that of the resected cartilage in the early surgery. Of 4 patients (1 from group and 3 from group 2), nothing on the donor site was found. From histologic evaluation, fibrocalcific tissue was seen, and cartilage cells were not seen in 2 patients with high CT value. CONCLUSION: Clinical observation presented that regenerative tissue at the donor site after harvesting costal cartilage, leaving the subjacent perichondrium completely intact, was mostly fibrocalcific tissue rather than cartilage tissue. The authors suspect that the perichondrium itself may not have regenerative power, but as an envelope for regeneration, perichondrium has a role.

Detection of Cassava Component in Sweet Potato Noodles by Real-Time Loop-mediated Isothermal Amplification (Real-time LAMP) Method.

Sweet potato (Ipomoea batatas) noodles are a traditional Chinese food with a high nutritional value; however, starch adulteration is a big concern. The objective of this study was to develop a reliable method for the rapid detection of cassava (Manihot esculenta) components in sweet potato noodles to protect consumers from commercial adulteration. Five specific Loop-mediated Isothermal Amplification (LAMP) primers targeting the internal transcribed spacer (ITS) of cassava were designed, genomic DNA was extracted, the LAMP reaction system was optimized, and the specificity of the primers was verified with genomic DNA of cassava, Ipomoea batatas, Zea mays, and Solanum tuberosum; the detection limit was determined with a serial dilution of adulterated sweet potato starch with cassava starch, and the real-time LAMP method for the detection of the cassava-derived ingredient in sweet potato noodles was established. The results showed that the real-time LAMP method can accurately and specifically detect the cassava component in sweet potato noodles with a detection limit of 1%. Furthermore, the LAMP assay was validated using commercial sweet potato noodle samples, and results showed that 57.7% of sweet potato noodle products (30/52) from retail markets were adulterated with cassava starch in China. This study provides a promising solution for facilitating the surveillance of the commercial adulteration of sweet potato noodles from retail markets.

The design of Fe, N-doped hierarchically porous carbons as highly active and durable electrocatalysts for a Zn-air battery.

A new type of Fe, N-doped hierarchically porous carbons (N-Fe-HPCs) has been synthesized via a cost-effective synthetic route, derived from nitrogen-enriched polyquaternium networks by combining a simple silicate templated two-step graphitization of the impregnated carbon. The as-prepared N-Fe-HPCs present a high catalytic activity for the oxygen reduction reaction (ORR) with onset and half-wave potentials of 0.99 and 0.86 V in 0.1 M KOH, respectively, which are superior to commercially available Pt/C catalyst (half-wave potential 0.86 V vs. RHE). Surprisingly, the diffusion-limited current density of N-S-HPCs approaches ∼7.5 mA cm(-2), much higher than that of Pt/C (∼5.5 mA cm(-2)). As a cathode electrode material used in Zn-air batteries, the unique configuration of the N-Fe-HPCs delivers a high discharge peak power density reaching up to 540 mW cm(-2) with a current density of 319 mA cm(-2) at 1.0 V of cell voltage and an energy density >800 Wh kg(-1). Additionally, outstanding ORR durability of the N-Fe-HPCs is demonstrated, as evaluated by the transient cell-voltage behavior of the Zn-air battery retaining an open circuit voltage of 1.48 V over 10 hours with a discharge current density of 100 mA cm(-2).

A Comparison of In-House Real-Time LAMP Assays with a Commercial Assay for the Detection of Pathogenic Bacteria.

Molecular detection of bacterial pathogens based on LAMP methods is a faster and simpler approach than conventional culture methods. Although different LAMP-based methods for pathogenic bacterial detection are available, a systematic comparison of these different LAMP assays has not been performed. In this paper, we compared 12 in-house real-time LAMP assays with a commercialized kit (Isothermal Master Mix) for the detection of Listeria monocytogenes, Salmonella spp, Staphylococcus aureus, Escherichia coli O157, E. coli O26, E. coli O45, E. coli O103, E. coli O111, E. coli O121, E. coli O145 and Streptococcus agalactiae. False-positive results were observed in all 12 in-house real-time LAMP assays, while all the negative controls of Isothermal Master Mix remained negative after amplification. The detection limit of Isothermal Master Mix for Listeria monocytogenes, Salmonella spp, Staphylococcus aureus, Escherichia coli O157, E. coli O26, E. coli O45, E. coli O103, E. coli O111, E. coli O121 and Streptococcus agalactiae was 1 pg, whereas the sensitivity of the commercialized kit for E. coli O145 was 100 pg. In conclusion, the 12 in-house real-time LAMP assays were impractical to use, while the commercialized kit Isothermal Master Mix was useful for the detection of most bacterial pathogens.

Efficient Ethane and Propane Separation from Natural Gas Using Heterometallic Metal-Organic Frameworks with Interpenetrated Structures.

The development of efficient technology for natural gas separation in industrial processes has become imperative. In this regard, the exploration of novel and effective adsorbents has gained significant attention. One promising approach is the metal regulation of metal-organic frameworks (MOFs), particularly heterometallic MOFs, which offer greater potential for gas separation due to their diverse composition. This study presents the synthesis of a series of iron- and vanadium-based heterometallic MOFs (MIL-126), featuring interpenetrated structures, and investigates their adsorption performance for methane (CH4), ethane (C2H6), and propane (C3H8). Experimental results reveal that the choice of metal combinations within the MOF framework significantly influences the adsorption performance of MIL-126. Notably, heterometallic MIL-126(Fe/Ni) exhibits a stronger binding affinity for C3H8, with an impressive uptake of 177 cm3/g. The C3H8/CH4 ideal adsorbed solution theory selectivity of MIL-126(Fe/Ni) surpasses that of MIL-126(Fe) by a factor of 7, reaching a value of 853, second only to the highest reported value. Furthermore, MIL-126(Fe/Ni) exhibits remarkable potential for the recovery of pure CH4 from the equimolar C3H8/CH4 mixture, with the amount of pure CH4 approaching the maximum reported value for MOFs. Insights from isosteric heat at zero loading and Henry's coefficients indicate that the transformation of metal types leads to a change in the interaction energy between C3H8 and the framework. Furthermore, breakthrough experiments validate the effective separation capability of MIL-126(Fe/Ni) for CH4/C2H6/C3H8 mixtures. These findings underscore the remarkable potential of heterometallic MOFs in constructing a wide range of new MOFs with tailorable properties, thereby enhancing their gas separation performance.

Portable hydroxyl-functionalized coal gangue-based cordierite porous ceramics sheets for effective adsorption of fluorine-containing wastewater.

Monolithic adsorbent removal of fluoride from water is considered an effective and non-secondary pollution method. Here, a portable hydroxyl-functionalized coal gangue-based cordierite porous ceramic sheet (ACGC-Fe) is prepared by using coal gangue solid waste with a specific silicon-aluminum-rich composition ratio and a small amount of magnesium oxide as a raw material through powder compression molding and mild chemical modification. The prepared ACGC-Fe can be used to treat fluorine-containing wastewater and the maximum adsorption of fluorine can reach 18.69 mg g-1. The Langmuir (Freundlich) adsorption isotherm model and pseudo-second-order kinetic model here provided a satisfactory description of the fluoride removal operating mechanism, and it is confirmed that the adsorption mechanism of ACGC-Fe is mainly attributed to the chemisorption of hydrogen bonds (with hydroxyl group) and ionic bonds (with metal), and physical adsorption based on cordierite porous ceramic pores. This research will provide a new idea for designing high-performance materials by mining and analyzing the composition and structure characteristics of coal gangue solid waste itself and broaden the application range of high-value-added coal gangue solid waste.

Molecular modelling studies reveal cryoprotective mechanism of L-Proline during the frozen storage of shrimp (Litopenaeus vannamei).

This study aimed to investigate the cryoprotective properties of proline (1% and 3% (w/v)) on shrimp. The cryoprotective mechanism was studied using physico-chemical experiments and molecular simulations. Proline had a notable positive impact on the thawing loss and texture of shrimp in comparison to the control. The denaturation of myosin in frozen shrimp was delayed by proline. Microscopy analysis demonstrated that proline effectively lowered the harm caused by ice crystals to shrimp muscle. Molecular simulations indicated that proline potentially exerted a cryoprotective effect primarily through the "water substitution" and "glassy state" hypotheses. Proline formed hydrogen bonds with myosin to replace the water molecules around myosin. Additionally, proline interacted with water molecules to form a glassy state, impeding the growth of ice crystals. Consequently, the stability of shrimp myosin was enhanced during freezing. In conclusion, proline demonstrated promise as an efficacious cryoprotectant for aquatic products.

UCL-Dehaze: Toward Real-World Image Dehazing via Unsupervised Contrastive Learning.

While the wisdom of training an image dehazing model on synthetic hazy data can alleviate the difficulty of collecting real-world hazy/clean image pairs, it brings the well-known domain shift problem. From a different yet new perspective, this paper explores contrastive learning with an adversarial training effort to leverage unpaired real-world hazy and clean images, thus alleviating the domain shift problem and enhancing the network's generalization ability in real-world scenarios. We propose an effective unsupervised contrastive learning paradigm for image dehazing, dubbed UCL-Dehaze. Unpaired real-world clean and hazy images are easily captured, and will serve as the important positive and negative samples respectively when training our UCL-Dehaze network. To train the network more effectively, we formulate a new self-contrastive perceptual loss function, which encourages the restored images to approach the positive samples and keep away from the negative samples in the embedding space. Besides the overall network architecture of UCL-Dehaze, adversarial training is utilized to align the distributions between the positive samples and the dehazed images. Compared with recent image dehazing works, UCL-Dehaze does not require paired data during training and utilizes unpaired positive/negative data to better enhance the dehazing performance. We conduct comprehensive experiments to evaluate our UCL-Dehaze and demonstrate its superiority over the state-of-the-arts, even only 1,800 unpaired real-world images are used to train our network. Source code is publicly available at https://github.com/yz-wang/UCL-Dehaze.

Ultrastiff metamaterials generated through a multilayer strategy and topology optimization.

Metamaterials composed of different geometrical primitives have different properties. Corresponding to the fundamental geometrical forms of line, plane, and surface, beam-, plate-, and shell-based lattice metamaterials enjoy many advantages in many aspects, respectively. To fully exploit the advantages of each structural archetype, we propose a multilayer strategy and topology optimization technique to design lattice metamaterial in this study. Under the frame of the multilayer strategy, the design space is enlarged and diversified, and the design freedom is increased. Topology optimization is applied to explore better designs in the larger and diverse design space. Beam-plate-shell-combined metamaterials automatically emerge from the optimization to achieve ultrahigh stiffness. Benefiting from high stiffness, energy absorption performances of optimized results also demonstrate substantial improvements under large geometrical deformation. The multilayer strategy and topology optimization can also bring a series of tunable dimensions for lattice design, which helps achieve desired mechanical properties, such as isotropic elasticity and functionally grading material property, and superior performances in acoustic tuning, electrostatic shielding, and fluid field tuning. We envision that a broad array of synthetic and composite metamaterials with unprecedented performance can be designed with the multilayer strategy and topology optimization.

Genome-wide association study and haplotype analysis reveal novel candidate genes for resistance to powdery mildew in soybean.

Powdery mildew disease (PMD) is caused by the obligate biotrophic fungus Microsphaera diffusa Cooke & Peck (M. diffusa) and results in significant yield losses in soybean (Glycine max (L.) Merr.) crops. By identifying disease-resistant genes and breeding soybean accessions with enhanced resistance, we can effectively mitigate the detrimental impact of PMD on soybeans. We analyzed PMD resistance in a diversity panel of 315 soybean accessions in two locations over 3 years, and candidate genes associated with PMD resistance were identified through genome-wide association studies (GWAS), haplotype analysis, qRT-PCR, and EMS mutant analysis. Based on the GWAS approach, we identified a region on chromosome 16 (Chr16) in which 21 genes form a gene cluster that is highly correlated with PMD resistance. In order to validate and refine these findings, we conducted haplotype analysis of 21 candidate genes and indicated there are single nucleotide polymorphisms (SNPs) and insertion-deletions (InDels) variations of Glyma.16G214000, Glyma.16G214200, Glyma.16G215100 and Glyma.16G215300 within the coding and promoter regions that exhibit a strong association with resistance against PMD. Subsequent structural analysis of candidate genes within this cluster revealed that in 315 accessions, the majority of accessions exhibited resistance to PMD when Glyma.16G214300, Glyma.16G214800 and Glyma.16G215000 were complete; however, they demonstrated susceptibility to PMD when these genes were incomplete. Quantitative real-time PCR assays (qRT-PCR) of possible candidate genes showed that 14 candidate genes (Glyma.16G213700, Glyma.16G213800, Glyma.16G213900, Glyma.16G214000, Glyma.16G214200, Glyma.16G214300, Glyma.16G214500, Glyma.16G214585, Glyma.16G214669, Glyma.16G214700, Glyma.16G214800, Glyma.16G215000, Glyma.16G215100 and Glyma.16G215300) were involved in PMD resistance. Finally, we evaluated the PMD resistance of mutant lines from the Williams 82 EMS mutations library, which revealed that mutants of Glyma.16G214000, Glyma.16G214200, Glyma.16G214300, Glyma.16G214800, Glyma.16G215000, Glyma.16G215100 and Glyma.16G215300, exhibited sensitivity to PMD. Combined with the analysis results of GWAS, haplotypes, qRT-PCR and mutants, the genes Glyma.16G214000, Glyma.16G214200, Glyma.16G214300, Glyma.16G214800, Glyma.16G215000, Glyma.16G215100 and Glyma.16G215300 were identified as highly correlated with PMD resistance. The candidate genes identified above are all NLR family genes, and these discoveries deepen our understanding of the molecular basis of PMD resistance in soybeans and will be useful for guiding breeding strategies.