Explore the vast range of titles available.

Aircraft recognition in remote sensing images has long been a meaningful topic. Most related methods treat entire images as a whole and do not concentrate on the features of parts. In fact, a variety of aircraft types have small interclass variance, and the main evidence for classifying subcategories is related to some discriminative object parts. In this paper, we introduce the idea of fine-grained visual classification (FGVC) and attempt to make full use of the features from discriminative object parts. First, multiple class activation mapping (MultiCAM) is proposed to extract the discriminative parts of aircrafts of different categories. Second, we present a mask filter (MF) strategy to enhance the discriminative object parts and filter the interference of the background from original images. Third, a selective connected feature fusion method is proposed to fuse the features extracted from both networks, focusing on the original images and the results of MF, respectively. Compared with the single prediction category in class activation mapping (CAM), MultiCAM makes full use of the predictions of all categories to overcome the wrong discriminative parts produced by a wrong single prediction category. Additionally, the designed MF preserves the object scale information and helps the network to concentrate on the object itself rather than the interfering background. Experiments on a challenging dataset prove that our method can achieve state-of-the-art performance.

The proteoglycan Syndecan-2 (Sdc2) has been implicated in regulation of cytoskeleton organization, integrin signaling and developmental angiogenesis in zebrafish. Here we report that mice with global and inducible endothelial-specific deletion of Sdc2 display marked angiogenic and arteriogenic defects and impaired VEGFA 165 signaling. No such abnormalities are observed in mice with deletion of the closely related Syndecan-4 (Sdc4) gene. These differences are due to a significantly higher 6-O sulfation level in Sdc2 versus Sdc4 heparan sulfate (HS) chains, leading to an increase in VEGFA 165 binding sites and formation of a ternary Sdc2-VEGFA 165 -VEGFR2 complex which enhances VEGFR2 activation. The increased Sdc2 HS chains 6-O sulfation is driven by a specific N-terminal domain sequence; the insertion of this sequence in Sdc4 N-terminal domain increases 6-O sulfation of its HS chains and promotes Sdc2-VEGFA 165 -VEGFR2 complex formation. This demonstrates the existence of core protein-determined HS sulfation patterns that regulate specific biological activities. Proteoglycans are glycosylated proteins that play a number of structural and signalling functions. Here, Corti, Wang et al. show that the N-terminal sequence of proteoglycan Syndecan-2 selectively increases 6-O sulfation of its heparan sulfate chains, and that this promotes formation of a ternary Sdc2/VEGFA/VEGFR2 complex leading to increased angiogenesis.

The O-antigen polysaccharide (O-PS) component of lipopolysaccharides on the surface of gram-negative bacteria is both a virulence factor and a B-cell antigen. Antibodies elicited by O-PS often confer protection against infection; therefore, O-PS glycoconjugate vaccines have proven useful against a number of different pathogenic bacteria. However, conventional methods for natural extraction or chemical synthesis of O-PS are technically demanding, inefficient, and expensive. Here, we describe an alternative methodology for producing glycoconjugate vaccines whereby recombinant O-PS biosynthesis is coordinated with vesiculation in laboratory strains of Escherichia coli to yield glycosylated outer membrane vesicles (glycOMVs) decorated with pathogen-mimetic glycotopes. Using this approach, glycOMVs corresponding to eight different pathogenic bacteria were generated. For example, expression of a 17-kb O-PS gene cluster from the highly virulent Francisella tularensis subsp. tularensis (type A) strain Schu S4 in hypervesiculating E. coli cells yielded glycOMVs that displayed F. tularensis O-PS. Immunization of BALB/c mice with glycOMVs elicited significant titers of O-PS–specific serum IgG antibodies as well as vaginal and bronchoalveolar IgA antibodies. Importantly, glycOMVs significantly prolonged survival upon subsequent challenge with F. tularensis Schu S4 and provided complete protection against challenge with two different F. tularensis subsp. holarctica (type B) live vaccine strains, thereby demonstrating the vaccine potential of glycOMVs. Given the ease with which recombinant glycotopes can be expressed on OMVs, the strategy described here could be readily adapted for developing vaccines against many other bacterial pathogens.

Collaborative perception enhances onboard perceptual capability by integrating features from other platforms, effectively mitigating the compromised accuracy caused by a restricted observational range and vulnerability to interference. However, current implementations of collaborative perception overlook the prevalent issues of both limited and low-reliability communication, as well as misaligned observations in remote sensing. To address this problem, this article presents an innovative distributed collaborative perception network (DCP-Net) specifically designed for remote sensing applications. Firstly, a self-mutual information match module is proposed to identify collaboration opportunities and select suitable partners. This module prioritizes critical collaborative features and reduces redundant transmission for better adaptation to weak communication in remote sensing. Secondly, a related feature fusion module is devised to tackle the misalignment between local and collaborative features due to the multiangle observations, improving the quality of fused features for the downstream task. We conduct extensive experiments and visualization analyses using three semantic segmentation datasets, namely Potsdam, iSAID, and DFC23. The results demonstrate that DCP-Net outperforms the existing collaborative perception methods comprehensively, improving mIoU by 2.61% to 16.89% at the highest collaboration efficiency and achieving state-of-the-art performance.

Predictable, clean genetic modification (GM) in livestock is important for reliable phenotyping and biosafety. Here we reported the generation of isozygous, functional myostatin (MSTN) knockout cloned pigs free of selectable marker gene (SMG) by CRISPR/Cas9 and Cre/LoxP. CRISPR/Cas9-mediated homologous recombination (HR) was exploited to knock out (KO) one allele of MSTN in pig primary cells. Cre recombinase was then used to excise the SMG with an efficiency of 82.7%. The SMG-free non-EGFP cells were isolated by flow cytometery and immediately used as donor nuclei for nuclear transfer. A total of 685 reconstructed embryos were transferred into three surrogates with one delivering two male live piglets. Molecular testing verified the mono-allelic MSTN KO and SMG deletion in these cloned pigs. Western blots showed approximately 50% decrease in MSTN and concurrent increased expression of myogenic genes in muscle. Histological examination revealed the enhanced myofiber quantity but myofiber size remained unaltered. Ultrasonic detection showed the increased longissimus muscle size and decreased backfat thickness. Precision editing of pig MSTN gene has generated isozygous, SMG-free MSTN KO cloned founders, which guaranteed a reliable route for elite livestock production and a strategy to minimize potential biological risks.

Unsupervised multi-source domain adaptation methods transfer knowledge learned from multiple labeled source domains to an unlabeled target domain. Existing methods assume that all source domain data can be accessed directly. However, such an assumption is unrealistic and causes data privacy concerns, especially when the source domain labels include personal information. In such a setting, it is prohibited to minimize domain gaps by pairwise calculation of the data from the source and target domains. Therefore, this work addresses the source-free unsupervised multi-source domain adaptation problem, where only the source models are available during the adaptation. We propose trust center sample-based source-free domain adaptation (TSDA) method to solve this problem. The key idea is to leverage the pre-trained models from the source domain and progressively train the target model in a self-learning manner. Because target samples with low entropy measured from the pre-trained source model achieve high accuracy, the trust center samples are selected first using the entropy function. Then pseudo labels are assigned for target samples based on a self-supervised pseudo-labeling strategy. For multiple source domains, corresponding target models are learned based on the assigned pseudo labels. Finally, multiple target models are integrated to predict the label for unlabeled target data. Extensive experiment results on some benchmark datasets and generated adversarial samples demonstrate that our approach outperforms existing UMDA methods, even though some methods can always access source data.

Background Bletilla striata is the main medicine of many skin whitening classic formulas in traditional Chinese medicine (TCM) and is widely used in cosmetic industry recently. However, its active ingredients are still unclear and its fibrous roots are not used effectively. The aim of the present study is to discover and identify its potential anti-melanogenic active constituents by zebrafish model and molecular docking. Methods The antioxidant activities were evaluated by 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity, 2,2′-azino-bis-(3-ethylbenthiazoline-6-sulphonic acid) (ABTS) radical scavenging activity and ferric reducing antioxidant power (FRAP) assay . The anti-melanogenic activity was assessed by tyrosinase inhibitory activity in vitro and melanin inhibitory in zebrafish. The chemical profiles were performed by ultra-high-performance liquid chromatography combined with quadrupole time-of-flight tandem mass spectrometry (UPLC-Q-TOF-MS/MS). Meanwhile, the potential anti-melanogenic active constituents were temporary identified by molecular docking. Results The 95% ethanol extract of B. striata fibrous roots (EFB) possessed the strongest DPPH, ABTS, FRAP and tyrosinase inhibitory activities, with IC 50 5.94 mg/L, 11.69 mg/L, 6.92 mmol FeSO 4 /g, and 58.92 mg/L, respectively. In addition, EFB and 95% ethanol extract of B. striata tuber (ETB) significantly reduced the melanin synthesis of zebrafish embryos in a dose-dependent manner. 39 chemical compositions, including 24 stilbenoids were tentatively identified from EFB and ETB. Molecular docking indicated that there were 83 (including 60 stilbenoids) and 85 (including 70 stilbenoids) compounds exhibited stronger binding affinities toward tyrosinase and adenylate cyclase. Conclusion The present findings supported the rationale for the use of EFB and ETB as natural skin-whitening agents in pharmaceutical and cosmetic industries.

PIWI-interacting RNAs (piRNAs) are a kind of important small non-coding RNAs and play a vital role in maintaining the stability of genome. Previous studies have revealed that piRNAs not only silence transposons, but also mediate the degradation of a large number of mRNAs and lncRNAs. Existing computational models only focus on mRNA-related piRNAs and rarely concentrate on lncRNA-related piRNAs. In this study, we propose a novel method, MLPPF, which is designed for multi-label prediction of piRNA functions based on pretrained k-mer, positional embedding and an improved TextRNN model. First, a benchmark dataset, which contains two types of functional labels, namely mRNA-related and lncRNA-related piRNAs, was constructed by processing piRNA-function-annotated data and sequence data. Moreover, pretrained k-mer embedding fused with positional embedding was applied to get the sequence representation with biological significance. Finally, an improved textRNN model with Bi-GRU and an attention mechanism was employed for implementing the piRNA functional label prediction task. Experiments substantiate that our model can effectively identify the piRNA functional labels, reveal the key factors of its subsequences and be helpful for in-depth investigations into piRNA functions.

In recent years, legged robots have been more and more widely used on non-structured terrain, and their foot structure has an important impact on the robot’s motion performance and stability. The structural characteristics of the yak foot sole with a high outer edge and low middle, which has excellent soil fixation ability and is an excellent bionic prototype, can improve the friction between the foot and the ground. At the same time, the foot hooves can effectively alleviate the larger impact load when contacting with the ground, which is an excellent anti-slip buffer mechanism. The bionic foot end design was carried out based on the morphology of the yak sole; the bionic foot design was carried out based on the biological anatomy observation of yak foot skeletal muscles. The virtual models of the bionic foot end and the bionic foot were established and simulated using Solidworks 2022 and Abaqus 2023, and the anti-slip performance on different ground surfaces and the influence of each parameter of the bionic foot on the cushioning effect were investigated. The results show that (1) the curved shape of the yak sole has a good anti-slip performance on both soil ground and rocky ground, and the anti-slip performance is better on soil ground; (2) the curved shape of the yak sole has a larger maximum static friction than the traditional foot, and the anti-slip performance is stronger under the same pressure conditions; (3) the finger pillow–hoof ball structure of the bionic foot has the greatest influence on the buffering effect, and the buffering effect of the bionic foot is best when the tip of the bionic foot touches the ground first.

Background No-tillage (NT) is a widely used field management to reduce soil erosion and degradation and is suggested to be beneficial for enhancing soil carbon (C) sequestration capacity. Nonetheless, the effects of NT on soil total carbon (TC) content in aeolian sandy soils are not extensively explored, and the underlying mechanisms are not clear. In our field experiments, the influence of NT and conventional tillage (CT) on sandy soil was studied. Methods We estimated the changes in soil TC in response to NT practice in a Cyperus esculentus L. field located at semi-arid Horqin sandy land, China. To unravel the underlying mechanisms, plant traits, soil properties and soil microbial characteristics were measured in parallel. The variations in soil bacterial community structure were investigated by 16S rRNA amplicon sequencing. The functionality of soil bacterial community was predicted based on OTU tables by using PICRUSt2. Results NT increased soil TC content in this sandy agroecosystem within a short-term experimental period, compared to CT. The underlying mechanisms might rely on three aspects. First, NT increased soil TC content through increasing photosynthesis and plant biomass, and thus, the plant-derived dissolved organic C. Second, NT increased the C immobilized in soil microbial biomass by increasing microbial C demands and C use efficiency. Third, NT increased the dominance of oligotrophic members in bacterial communities by decreasing available nutrient levels, which is associated with the recalcitrance and stability of the soil organic carbon. Conclusions The present study enriched our knowledge on the changes in the plant-soil-microbe continuum in response to NT in a semi-arid sandy agroecosystem. Still, this study provides a reference for modifying tillage practices to benefit crop yield as well as soil C sequestration.