Search

Hypoxic-ischemic encephalopathy (HIE) is a brain injury condition that poses a significant risk to newborns, potentially causing varying degrees of damage to the central nervous system. Its clinical manifestations include respiratory distress, cardiac dysfunction, hypotension, muscle weakness, seizures, and coma. As HIE represents a progressive brain injury, early identification of the extent of the damage and the implementation of appropriate treatment are crucial for reducing mortality and improving outcomes. HIE patients may face long-term complications such as cerebral palsy, epilepsy, vision loss, and developmental delays. Therefore, prompt identification and treatment of hypoxic-ischemic symptoms can help reduce the risk of severe sequelae in patients. Currently, hypothermia therapy is one of the most effective treatments for HIE patients. However, not all newborns with HIE are suitable for this therapy, making rapid and accurate assessment of the extent of brain injury critical for treatment. Among HIE patients, hypothermia therapy has shown better efficacy in those diagnosed with moderate to severe HIE within 6 h of birth, establishing this time frame as the golden period for treatment. During this golden period, an accurate assessment of HIE severity is essential for formulating appropriate treatment strategies and predicting long-term outcomes for the affected infants. This study proposes a method for addressing data imbalance and noise interference through data preprocessing techniques, including filtering and SMOTE. It then employs EEGNet, a deep learning model specifically designed for EEG classification, combined with a Transformer model featuring an attention mechanism that excels at capturing long-term sequential features to construct the Trans-EEGNet model. This model outperforms previous methods in computation time and feature extraction, enabling rapid classification and assessment of HIE severity in newborns.

Weeds significantly impact paddy yields, and herbicides offer a cost-effective, rapid, and efficient solution compared to manual weeding, ensuring agricultural productivity. Tripyrasulfone, a novel 4-hydroxyphenylpyruvate dioxygenase (HPPD) inhibitor developed by Qingdao Kingagroot Chemicals Co., Ltd., has demonstrated high efficacy when applied post-emergence, causing characteristic foliar bleaching in susceptible weed species, distinct from conventional acetolactate synthase, acetyl-CoA carboxylase, and synthetic auxin herbicides. This study investigates the impact of tripyrasulfone on the activity of key soil enzymes (urease (UE), acid phosphatase (ACP), sucrase (SC), catalase (CAT), and dehydrogenase (DHA)) in paddy soils from Jilin Province and Shandong Province. Different doses of tripyrasulfone (0.1, 1.0, and 2.5 mg kg−1) were applied, and the enzymatic activities were measured. Results indicated that tripyrasulfone initially inhibited UE and ACP activities before activating them. On the 20th day after treatment, UE activity had returned to control levels, whereas ACP activity remained significantly higher, showing long-lasting activation. SC and CAT activities were inhibited but gradually recovered to control levels. Furthermore, DHA activity was activated with a sustained effect, remaining significantly higher than the control group even 20 days after treatment. Overall, the impact of tripyrasulfone on soil enzyme activities diminished over time, suggesting that tripyrasulfone posed minimal long-term ecological risk to soil health.

This study fabricated piezoelectric fibers of polyvinylidene fluoride (PVDF) with graphene using near-field electrospinning (NFES) technology. A uniform experimental design table U*774 was applied, considering weight percentage (1–13 wt%), the distance between needle and disk collector (2.1–3.9 mm), and applied voltage (14.5–17.5 kV). We optimized the parameters using electrical property measurements and the Kriging response surface method. Adding 13 wt% graphene significantly improved electrical conductivity, increasing from 17.7 µS/cm for pure PVDF to 187.5 µS/cm. The fiber diameter decreased from 21.4 µm in PVDF/1% graphene to 9.1 µm in PVDF/13% graphene. Adding 5 wt% graphene increased the β-phase content by 6.9%, reaching 65.4% compared to pure PVDF fibers. Material characteristics were investigated using scanning electron microscopy (SEM), Fourier-transform infrared spectroscopy (FTIR), X-ray diffraction analysis (XRD), contact angle measurements, and tensile testing. Optimal parameters included 3.47 wt% graphene, yielding 15.82 mV voltage at 5 Hz and 5 N force (2.04 times pure PVDF). Force testing showed a sensitivity (S) of 7.67 log(mV/N). Fibers were attached to electrodes for piezoelectric sensor applications. The results affirmed enhanced electrical conductivity, piezoelectric performance, and mechanical strength. The optimized piezoelectric sensor could be applied to measure physiological signals, such as attaching it to the throat under different conditions to measure the output voltage. The force-to-voltage conversion facilitated subsequent analysis.

Mechanical force is the basis of cardiovascular development, homeostasis, and diseases. The perception and response of mechanical force by the cardiovascular system are crucial. However, the molecular mechanisms mediating mechanotransduction in the cardiovascular system are not yet understood. PIEZO1, a novel transmembrane mechanosensitive cation channel known for its regulation of touch sensation, has been found to be widely expressed in the mammalian cardiovascular system. In this review, we elucidate the role and mechanism of PIEZO1 as a mechanical sensor in cardiovascular development, homeostasis, and disease processes, including embryo survival, angiogenesis, cardiac development repair, vascular inflammation, lymphangiogenesis, blood pressure regulation, cardiac hypertrophy, cardiac fibrosis, ventricular remodeling, and heart failure. We further summarize chemical molecules targeting PIEZO1 for potential translational applications. Finally, we address the controversies surrounding emergent concepts and challenges in future applications.

Knowledge graphs (KGs) serve as structured representations of knowledge, comprising entities and relations. KGs are inherently incomplete, sparse, and have a strong need for completion. Although many knowledge graph embedding models have been designed for knowledge graph completion, they predominantly focus on capturing observable correlations between entities. Due to the sparsity of KGs, potential semantic correlations are challenging to capture. To tackle this problem, we propose a model entitled semantic-enhanced knowledge graph completion (SE-KGC). SE-KGC effectively addresses the issue by incorporating predefined semantic patterns, enabling the capture of semantic correlations between entities and enhancing features for representation learning. To implement this approach, we employ a multi-relational graph convolution network encoder, which effectively encodes the KG. Subsequently, we utilize a scoring decoder to evaluate triplets. Experimental results demonstrate that our SE-KGC model outperforms other state-of-the-art methods in link-prediction tasks across three datasets. Specifically, compared to the baselines, SE-KGC achieved improvements of 11.7%, 1.05%, and 2.30% in terms of MRR on these three datasets. Furthermore, we present a comprehensive analysis of the contributions of different semantic patterns, and find that entities with higher connectivity play a pivotal role in effectively capturing and characterizing semantic information.

For the purpose of investigating the microstructure deformation of 28Mn-10Al-C steel at high speeds under different strain rates, the dynamic properties of 28Mn-10Al-C steel under varying strain rates and the feasibility of the tensile specimens with a variable cross-section were evaluated using a combination of tensile test, optical microscopy (OM), X-ray diffraction (XRD), scanning electron microscope (SEM), and electron back scatter diffraction (EBSD). The test results demonstrate that the high-tension tensile process of 28Mn-10Al-C steel involves a competitive process of work hardening, deformation speed reinforcement, and adiabatic temperature elevation. The elasticity limit, tensile strength, and elongation of 28Mn-10Al-C steel increase with the rate of deformation. Specifically, at a deformation rate of 103 s−1, the yield strength, tensile strength, and elongation of the test steel increase to 817 MPa, 1047 MPa, and 60.6%, respectively, indicating significant improvements in all properties. Through analyzing its mechanical properties, dislocation density, and angle grain boundary density, this article discusses the deformation behavior of 28Mn-10Al-C steel during dynamic deformation. It is found that the dominant hardening mechanism and softening mechanism in the deformation process change with the increase in strain rate.

It is known that a large amount of vibration mechanical energy will be generated during train operation. If the mechanical energy can be obtained from the track structure, it can greatly optimize the energy configuration of the metro. Currently, most sensors are limited to disk or cymbal structures and are not used in the track bed; different from existing structures, this paper designs a spherical energy harvester based on a rubber bearing floating slab track, wherein the size range of the spherical energy harvesting structure was approximately determined based on the geometric spatial relationships of the actual track bed internal structure. Compared to the traditional disk and cymbal energy harvesters, the mechanical and electrical properties of the spherical energy harvesting structure was studied by a numerical simulation method, and the optimal size of the spherical energy was determined by calculation. The main conclusions are as follows: (1) Compared with the traditional disk harvester and cymbal harvester, the spherical harvester had better mechanical and electrical properties. (2) By calculating the output power of energy harvesters under load matching, we found that the output power of the spherical harvester was two orders of magnitude higher than that of the disk harvester and 53% higher than that of the cymbal harvester. (3) The optimum size of the spherical harvester was when the thickness of piezoelectric layer was 2 mm, the radius was 16 mm, the axial ratio of the spherical shell was 0.4, and the thickness of the spherical shell was 4 mm.

Research on government service quality can help ensure the success of digital government services and has been the focus of numerous studies that proposed different frameworks and approaches. Most of the existing studies are based on traditional researcher-led methods, which struggle to capture the needs of citizens. In this paper, a citizen-feedback-based analysis framework was proposed to explore citizen demands and analyze the service quality of digital government. Citizen feedback data are a direct expression of citizens’ demands, so the citizen-feedback-based framework can help to obtain more targeted management insights and improve citizen satisfaction. Efficient machine learning methods used in the framework make data collection and processing more efficient, especially for large-scale internet data. With the crawled user feedback data from the Q&A e-government portal of Luzhou, Sichuan Province, China, we conducted experiments on the proposed framework to verify its feasibility. From citizens’ online feedback on Q&A services, we extracted five service quality factors: efficiency, quality, attitude, compliance, and execution of response. The analysis of five service quality factors provides some management insights, which can provide a guide for improvements in Q&A services.

Botulinum neurotoxins (BoNTs) can cause nerve paralysis syndrome in mammals and other vertebrates. BoNTs are the most toxic biotoxins known and are classified as Class A biological warfare agents. BoNTs are mainly divided into seven serotypes A-G and new neurotoxins BoNT/H and BoNT/X, which have similar functions. BoNT proteins are 150 kDa polypeptide consisting of two chains and three domains: the light chain (L, catalytic domain, 50 kDa) and the heavy chain (H, 100 kDa), which can be divided into an N-terminal membrane translocation domain (HN, 50 kDa) and a C-terminal receptor binding domain (Hc, 50 kDa). In current study, we explored the immunoprotective efficacy of each functional molecule of BoNT/F and the biological characteristics of the light chain-heavy N-terminal domain (FL-HN). The two structure forms of FL-HN (i.e., FL-HN-SC: single chain FL-HN and FL-HN-DC: di-chain FL-HN) were developed and identified. FL-HN-SC could cleave the vesicle associated membrane protein 2 (VAMP2) substrate protein in vitro as FL-HN-DC or FL. While only FL-HN-DC had neurotoxicity and could enter neuro-2a cells to cleave VAMP2. Our results showed that the FL-HN-SC had a better immune protection effect than the Hc of BoNT/F (FHc), which indicated that L-HN-SC, as an antigen, provided the strongest protective effects against BoNT/F among all the tested functional molecules. Further in-depth research on the different molecular forms of FL-HN suggested that there were some important antibody epitopes at the L-HN junction of BoNT/F. Thus, FL-HN-SC could be used as a subunit vaccine to replace the FHc subunit vaccine and/or toxoid vaccine, and to develop antibody immune molecules targeting L and HN domains rather than the FHc domain. FL-HN-DC could be used as a new functional molecule to evaluate and explore the structure and activity of toxin molecules. Further exploration of the biological activity and molecular mechanism of the functional FL-HN or BoNT/F is warranted.

α-conotoxin AuIB is the only one of the 4/6 type α-conotoxins (α-CTxs) that inhibits the γ-aminobutyric acid receptor B (GABABR)-coupled N-type calcium channel (CaV2.2). To improve its inhibitory activity, a series of variants were synthesized and evaluated according to the structure–activity relationships of 4/7 type α-CTxs targeting GABABR-coupled CaV2.2. Surprisingly, only the substitution of Pro7 with Arg results in a 2–3-fold increase in the inhibition of GABABR-coupled CaV2.2 (IC50 is 0.74 nM); substitutions of position 9–12 with basic or hydrophobic amino acid and the addition of hydrophobic amino acid Leu or Ile at the second loop to mimic 4/7 type α-CTxs all failed to improve the inhibitory activity of AuIB against GABABR-coupled CaV2.2. Interestingly, the most potent form of AuIB[P7R] has disulfide bridges of “1–4, 2–3” (ribbon), which differs from the “1–3, 2–4” (globular) in the isoforms of wildtype AuIB. In addition, AuIB[P7R](globular) displays potent analgesic activity in the acetic acid writhing model and the partial sciatic nerve injury (PNL) model. Our study demonstrated that 4/6 type α-CTxs, with the disulfide bridge connectivity “1–4, 2–3,” are also potent inhibitors for GABABR-coupled CaV2.2, exhibiting potent analgesic activity.

Graph contrastive learning (GCL) has been subject to more attention and been widely applied to numerous graph learning tasks such as node classification and link prediction. Although it has achieved great success and even performed better than supervised methods in some tasks, most of them depend on node-level comparison, while ignoring the rich semantic information contained in graph topology, especially for social networks. However, a higher-level comparison requires subgraph construction and encoding, which remain unsolved. To address this problem, we propose a subgraph adaptive structure-aware graph contrastive learning method (PASCAL) in this work, which is a subgraph-level GCL method. In PASCAL, we construct subgraphs by merging all motifs that contain the target node. Then we encode them on the basis of motif number distribution to capture the rich information hidden in subgraphs. By incorporating motif information, PASCAL can capture richer semantic information hidden in local structures compared with other GCL methods. Extensive experiments on six benchmark datasets show that PASCAL outperforms state-of-art graph contrastive learning and supervised methods in most cases.

Predictive graph learning approaches have been bringing significant advantages in many real-life applications, such as social networks, recommender systems, and other social-related downstream tasks. For those applications, learning models should be able to produce a great prediction result to maximize the usability of their application. However, the paradigm of current graph learning methods generally neglects the differences in link strength, leading to discriminative predictive results, resulting in different performance between tasks. Based on that problem, a fairness-aware predictive learning model is needed to balance the link strength differences and not only consider how to formulate it. To address this problem, we first formally define two biases (i.e., Preference and Favoritism) that widely exist in previous representation learning models. Then, we employ modularity maximization to distinguish strong and weak links from the quantitative perspective. Eventually, we propose a novel predictive learning framework entitled ACE that first implements the link strength differentiated learning process and then integrates it with a dual propagation process. The effectiveness and fairness of our proposed ACE have been verified on four real-world social networks. Compared to nine different state-of-the-art methods, ACE and its variants show better performance. The ACE framework can better reconstruct networks, thus also providing a high possibility of resolving misinformation in graph-structured data.

Relations in knowledge graphs have rich relational structures and various binary relational patterns. Various relation modelling strategies are proposed for embedding knowledge graphs, but they fail to fully capture both features of relations, rich relational structures and various binary relational patterns. To address the problem of insufficient embedding due to the complexity of the relations, we propose a novel knowledge graph representation model in complex space, namely MARS, to exploit complex relations to embed knowledge graphs. MARS takes the mechanisms of complex numbers and message-passing and then embeds triplets into relation-specific complex hyperplanes. Thus, MARS can well preserve various relation patterns, as well as structural information in knowledge graphs. In addition, we find that the scores generated from the score function approximate a Gaussian distribution. The scores in the tail cannot effectively represent triplets. To address this particular issue and improve the precision of embeddings, we use the standard deviation to limit the dispersion of the score distribution, resulting in more accurate embeddings of triplets. Comprehensive experiments on multiple benchmarks demonstrate that our model significantly outperforms existing state-of-the-art models for link prediction and triple classification.

Graph neural networks (GNNs) have demonstrated a remarkable ability in the task of semi-supervised node classification. However, most existing GNNs suffer from the nonrobustness issues, which poses a great challenge for applying GNNs into sensitive scenarios. Some researchers concentrate on constructing an ensemble model to mitigate the nonrobustness issues. Nevertheless, these methods ignore the interaction among base models, leading to similar graph representations. Moreover, due to the deterministic propagation applied in most existing GNNs, each node highly relies on its neighbors, leaving the nodes to be sensitive to perturbations. Therefore, in this paper, we propose a novel framework of graph ensemble learning based on knowledge passing (called GEL) to address the above issues. In order to achieve interaction, we consider the predictions of prior models as knowledge to obtain more reliable predictions. Moreover, we design a multilayer DropNode propagation strategy to reduce each node’s dependence on particular neighbors. This strategy also empowers each node to aggregate information from diverse neighbors, alleviating oversmoothing issues. We conduct experiments on three benchmark datasets, including Cora, Citeseer, and Pubmed. GEL outperforms GCN by more than 5% in terms of accuracy across all three datasets and also performs better than other state-of-the-art baselines. Extensive experimental results also show that the GEL alleviates the nonrobustness and oversmoothing issues.

This paper reports on a sensor based on multi-element complementary split-ring resonator for the measurement of liquid materials. The resonator consists of three split rings for improved measurement sensitivity. A hole is fabricated at the centre of the rings to accommodate a hollow glass tube, through which the liquid sample can be injected. Electromagnetic simulations demonstrate that both the resonant frequency and quality factor of the sensor vary considerably with the dielectric constant and loss tangent of the liquid sample. The volume ratio between the liquid sample and glass tube is 0.36, yielding great sensitivity in the measured results for high loss liquids. Compared to the design based on rectangular split rings, the proposed ring structure offers 37% larger frequency shifts and 9.1% greater resonant dips. The relationship between dielectric constant, loss tangent, measured quality factor and resonant frequency is derived. Experimental verification is conducted using ethanol solution with different concentrations. The measurement accuracy is calculated to be within 2.8%, and this validates the proposed approach.

α-Conotoxins GI and MI belong to the 3/5 subfamily of α-conotoxins and potently inhibit muscular nicotinic acetylcholine receptors (nAChRs). To date, no 3/4- or 3/6-subfamily α-conotoxins have been reported to inhibit muscular nAChRs. In the present study, a series of new 3/4-, 3/6-, and 3/7-subfamily GI and MI variants were synthesized and functionally characterized by modifications of loop2. The results show that the 3/4-subfamily GI variant GI[∆8G]-II and the 3/6-subfamily variants GI[+13A], GI[+13R], and GI[+13K] displayed potent inhibition of muscular nAChRs expressed in Xenopus oocytes, with an IC50 of 45.4–73.4 nM, similar to or slightly lower than that of wild-type GI (42.0 nM). The toxicity of these GI variants in mice appeared to be about a half to a quarter of that of wild-type GI. At the same time, the 3/7-subfamily GI variants showed significantly lower in vitro potency and toxicity. On the other hand, similar to the 3/6-subfamily GI variants, the 3/6-subfamily MI variants MI[+14R] and MI[+14K] were also active after the addition of a basic amino acid, Arg or Lys, in loop2, but the activity was not maintained for the 3/4-subfamily MI variant MI[∆9G]. Interestingly, the disulfide bond connectivity “C1–C4, C2–C3” in the 3/4-subfamily variant GI[∆8G]-II was significantly more potent than the “C1–C3, C2–C4” connectivity found in wild-type GI and MI, suggesting that disulfide bond connectivity is easily affected in the rigid 3/4-subfamily α-conotoxins and that the disulfide bonds significantly impact the variants’ function. This work is the first to demonstrate that 3/4- and 3/6-subfamily α-conotoxins potently inhibit muscular nAChRs, expanding our knowledge of α-conotoxins and providing new motifs for their further modifications.

N-methyl-D-aspartate receptor (NMDAR) antagonists have been found to be effective to inhibit morphine dependence. However, the discovery of the selective antagonist for NMDAR GluN2B with low side-effects still remains challenging. In the present study, we report a selective NMDAR GluN2B antagonist con-T[M8Q](a conantokin-T variant) that potently inhibits the naloxone-induced jumping and conditioned place preference of morphine-dependent mice at nmol/kg level, 100-fold higher than ifenprodil, a classical NMDAR NR2B antagonist. Con-T[M8Q] displays no significant impacts on coordinated locomotion function, spontaneous locomotor activity, and spatial memory mice motor function at the dose used. Further molecular mechanism experiments demonstrate that con-T[M8Q] effectively inhibited the transcription and expression levels of signaling molecules related to NMDAR NR2B subunit in hippocampus, including NR2B, p-NR2B, CaMKII-α, CaMKII-β, CaMKIV, pERK, and c-fos. The high efficacy and low side effects of con-T[M8Q] make it a good lead compound for the treatment of opiate dependence and for the reduction of morphine usage.

As the first in a new class of non-opioid drugs, ω-Conotoxin MVIIA was approved for the management of severe chronic pains in patients who are unresponsive to opioid therapy. Unfortunately, clinical application of MVIIA is severely limited due to its poor ability to penetrate the blood-brain barrier (BBB), reaching the central nervous system (CNS). In the present study, we have attempted to increase MVIIA’s ability to cross the BBB via a fusion protein strategy. Our results showed that when the TAT-transducing domain was fused to the MVIIA C-terminal with a linker of varied numbers of glycine, the MVIIA-TAT fusion peptide exhibited remarkable ability to cross the bio-membranes. Most importantly, both intravenous and intranasal administrations of MVIIA-TAT in vivo showed therapeutic efficacy of analgesia. Compared to the analgesic effects of intracerebral administration of the nascent MVIIA, these systemic administrations of MVIIA-TAT require higher doses, but have much prolonged effects. Taken together, our results showed that TAT conjugation of MVIIA not only enables its peripheral administration, but also maintains its analgesic efficiency with a prolonged effective time window. Intranasal administration also rendered the MVIIA-TAT advantages of easy applications with potentially reduced side effects. Our results may present an alternative strategy to improve the CNS accessibility for neural active peptides.

α-conotoxin GI, a short peptide toxin in the venom of Conus geographus, is composed of 13 amino acids and two disulfide bonds. It is the most toxic component of Conus geographus venom with estimated lethal doses of 0.029–0.038 mg/kg for humans. There is currently no reported analytical method for this toxin. In the present study, a sensitive detection method was developed to quantify GI in human plasma using a solid-phase extraction (SPE) column (polystyrene–divinyl benzene copolymer) combined with liquid chromatography/electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) in the multiple reaction monitoring (MRM) mode. The plasma samples were treated with a protein precipitating solvent (methanol: acetonitrile = 50:50, v/v). GI in the solvent was efficiently extracted with an SPE column and was further separated by a Grace Alltima HP C18 (50 × 2.1 mm, 5 μm) column at a flow rate of 0.4 mL/min. Water (with 2% methanol) acetonitrile (with 0.1% acetic acid) was selected as the mobile phase combination used in a linear gradient system. α-Conotoxin GI was analyzed by an API 4000 triple quadrupole mass spectrometer. In the method validation, the linear calibration curve in the range of 2.0 to 300.0 ng/mL had correlation coefficients (r) above 0.996. The recovery was 57.6–66.8% for GI and the internal standard. The lower limit of quantification (LLOQ) was 2 ng/mL. The intra- and inter-batch precisions were below 6.31% and 8.61%, respectively, and the accuracies were all within acceptance. GI was stable in a bench-top autosampler through long-term storage and freeze/thaw cycles. Therefore, this method is specific, sensitive and reliable for quantitative analysis of α-conotoxin GI in human plasma.