[Blood cadmium level and the change of cadmium exposure risk in Chinese urban school-age children in 2002 and 2012].

OBJECTIVE: To investigate the blood cadmium concentrations and the related change in Chinese urban children derived from the China Nutrition and Health Survey 2002 and 2012(CNHS 2002 and CNHS 2012). METHODS: The Chinese urban children aged 6-11 years were selected according to gender, age and regional distribution using the multi-stage stratified cluster random sampling method, as well as the corresponding whole blood samples. The blood cadmium concentration was carefully determined by the quadrupole inductively coupled plasma mass spectrometry(ICP-MS) and the percentage of blood cadmium over 2 µg/L was subsequently estimated. In addition, the upper limit values of the 95%CI of the 95th percentiles of available blood cadmium data was assessed as the threshold of cadmium exposure. RESULTS: Totally, 2182 Chinese urban children were included, and of these, 1036 children were from the CNHS 2002 and 1146 children were from the CNHS 2012. From the CNHS 2002 to the CNHS 2012, the median blood cadmium concentration was increased from 0.28 µg/L to 0.95 µg/L, and the percentage of blood cadmium with over 2 µg/L was elevated from 1.45% to 10.47%. In addition, the new estimated threshold of blood cadmium was ascended from 1.24 µg/L up to 2.89 µg/L. CONCLUSION: The risk of cadmium exposure in Chinese urban children aged 6-11 years was increasingly aggravated from the CHNS 2002 to the CNHS 2012.

The multi-year contribution of Indo-China peninsula fire emissions to aerosol radiation forcing in southern China during 2013-2019.

Fire emissions in Southeast Asia transported to southern China every spring (March-May), influencing not only the air quality but also the weather and climate. However, the multi-year variations and magnitude of this impact on aerosol radiation forcing in southern China remain unclear. Here, we quantified the multi-year contributions of fire emissions in Indo-China Peninsula (ICP) region to aerosol radiation forcing in the various southern Chinese provinces during the fire season (March-May) of 2013-2019 combining the 3-dimension chemical transport model and the Column Radiation Model (CRM) simulations. The models' evaluations showed they reasonably capture the temporal and spatial distribution of surface aerosol concentrations and column aerosol optical properties over the study regions. The fire emissions over the ICP region were found to increase the aerosol optical depth (AOD) value by 0.1 (15 %) and reduce the single scattering albedo (SSA) in three southern regions of China (Yunnan-YN, Guangxi-GX, and Guangdong-GD from west to east), owing to increases in the proportions of black carbon (BC, 0.4 % ± 0.1 %) and organic carbon (OC, 3.0 % ± 0.9 %) within the aerosol compositions. The transported smoke aerosols cooled surface but heated the atmosphere in the southern China regions, with the largest mean reduction of -5 Wm-2 (-3 %) in surface shortwave radiation forcing and the maximum daily contributions of about -15 Wm-2 (-15 %) to the atmosphere radiation forcing in the GX region, followed by the GD and YN regions. The impacts of ICP fire emissions on aerosol optical and radiative parameters declined during 2013-2019, with the highest rate of 0.393 ± 0.478 Wm-2 yr-1 in the GX for the shortwave radiation forcing in the atmosphere. Besides, their yearly changes in the contribution were consistent with the annual fire emissions in the ICP region. Such strong radiative perturbations of ICP fire emissions were expected to influence regional meteorology in southern China and should be considered in the climate simulations.

PSAT1 enhances the efficacy of the prognosis estimation nomogram model in stage-based clear cell renal cell carcinoma.

BACKGROUND: Clear cell renal cell carcinoma (ccRCC) is associated with a high prevalence of cancer-related deaths. The survival rates of patients are significantly lower in late-stage ccRCC than in early-stage ccRCC, due to the spread and metastasis of late-stage ccRCC, surgery has not reached the goal of radical cure, and the effect of traditional radiotherapy and chemotherapy is poor. Thus, it is crucial to accurately assess the prognosis and provide personalized treatment at an early stage in ccRCC. This study aims to develop an efficient nomogram model for stratifying and predicting the survival of ccRCC patients based on tumor stage. METHODS: We first analyzed the microarray expression data of ccRCC patients from the Gene Expression Omnibus (GEO) database and categorized them into two groups based on the disease stage (early and late stage). Subsequently, the GEO2R tool was applied to screen out the genes that were highly expressed in all GEO datasets. Finally, the clinicopathological data of the two patient groups were obtained from The Cancer Genome Atlas (TCGA) database, and the differences were compared between groups. Survival analysis was performed to evaluate the prognostic value of candidate genes (PSAT1, PRAME, and KDELR3) in ccRCC patients. Based on the screened gene PSAT1 and clinical parameters that were significantly associated with patient prognosis, we established a new nomogram model, which was further optimized to a single clinical variable-based model. The expression level of PSAT1 in ccRCC tissues was further verified by qRT-PCR, Western blotting, and immunohistochemical analysis. RESULTS: The datasets GSE73731, GSE89563, and GSE150404 identified a total of 22, 89, and 120 over-expressed differentially expressed genes (DEGs), respectively. Among these profiles, there were three genes that appeared in all three datasets based on different stage groups. The overall survival (OS) of late-stage patients was significantly shorter than that of early-stage patients. Among the three candidate genes (PSAT1, PRAME, and KDELR3), PSAT1 was shown to be associated with the OS of patients with late-stage ccRCC. Multivariate Cox regression analysis showed that age, tumor grade, neoadjuvant therapy, and PSAT1 level were significantly associated with patient prognosis. The concordance indices were 0.758 and 0.725 for the 3-year and 5-year OS, respectively. The new model demonstrated superior discrimination and calibration compared with the single clinical variable model. The enhancer PSAT1 used in the new model was shown to be significantly overexpressed in tissues from patients with late-stage ccRCC, as demonstrated by the mRNA level, protein level, and pathological evaluation. CONCLUSION: The new prognostic prediction nomogram model of PSAT1 and clinicopathological variables combined was thus established, which may provide a new direction for individualized treatment for different-stage ccRCC patients.

Effect of Sarcopenia on 10-Year Risk of Atherosclerotic Cardiovascular Disease in Patients with Type 2 Diabetes Mellitus.

Objective: To investigate the impact of sarcopenia on the 10-year risk of atherosclerotic cardiovascular disease (ASCVD) among individuals with type 2 diabetes mellitus (T2DM). Methods: This study included the clinical, laboratory, and body composition data of 1491 patients with T2DM who were admitted to the Department of Endocrinology and Metabolism at Tianjin Union Medical Center from July 2018 to July 2023. The China-PAR model was utilized to evaluate cardiovascular disease risk. Associations between ASCVD risk and various clinical parameters were analyzed, and the relationship between body composition parameters and ASCVD risk was assessed using logistic regression. Results: The analysis revealed that T2DM patients with sarcopenia had a higher 10-year ASCVD risk compared to those without sarcopenia, with reduced muscle mass independently predicting an increased risk of cardiovascular disease. This association was significant among female T2DM patients, while male T2DM patients with sarcopenia showed a marginally higher median ASCVD risk compared to their non-sarcopenic counterparts. ASCVD risk inversely correlated with body muscle parameters and positively correlated with fat content parameters. Specifically, height- and weight-adjusted fat mass (FM, FM%, FMI) were identified as risk factors for ASCVD. Conversely, muscle parameters adjusted for weight and fat (ASM%, SMM%, FFM%, ASM/FM, SMM/FM, FMM/FM) were protective against ASCVD risk. These findings highlight the critical role of sarcopenia in influencing cardiovascular disease risk among Chinese patients with T2DM, as predicted by the China-PAR model. Conclusion: This study highlights the importance of sarcopenia in T2DM patients, not only as an indicator of ASCVD risk, but possibly as an independent risk factor in this demographics.

Characterization of a novel VenusX orthogonal dual-layer multileaf collimator.

PURPOSE: To investigate and characterize the performance of a novel orthogonal dual-layer alpha multileaf collimator (αMLC) mounted on the LinaTech VenusX linac. METHODS: We evaluated leaf positioning accuracy and reproducibility using an electronic portal imaging device through the picket fence test. The average, interleaf, intraleaf, and leaf tip transmissions of the single and dual layers were measured using an ionization chamber. Square and rhombus fields were used to evaluate the leaf penumbra of αMLC. To investigate the advantages of the orthogonal dual-layer multileaf collimator (MLC) in field shaping, right triangular and circular pattern fields were formed using both the dual layers and single layers of the αMLC. RESULTS: The average maximum positioning deviations of the upper and lower αMLC over 1 year were 0.76 ± 0.09 mm and 0.62 ± 0.07 mm, respectively. The average transmissions were 1.87%, 1.83%, and 0.03% for the upper-, lower- and dual-layer αMLC, respectively. The maximum interleaf transmissions of the lower- and dual-layer were 2.43% and 0.17%, respectively. The leaf tip transmissions were 9.34% and 0.25%, respectively. The penumbra of the square field was 6.2 mm in the X direction and 8.0 mm in the Y direction. The average penumbras of the rhombus fields with side lengths of 5 and 10 cm were 3.6 and 4.9 mm, respectively. For the right triangular and circular fields, the fields shaped by the dual-layer leaves were much closer to the set field than those shaped by single-layer leaves. The dose undulation amplitude of the 50% isodose lines and leaf stepping angle change of the dual-layer leaves were smaller than those of the single-layer leaves. CONCLUSIONS: The αMLC benefits from its orthogonal dual-layer design. Leaf transmission, dose undulations at the field edge, and MLC field dependence of the leaf stepping angle of the dual-layer αMLC were remarkably reduced.

Terahertz-triggered ultrafast nonlinear optical activities in two-dimensional centrosymmetric PtSe2.

The nonlinear mechanisms of polarization and optical fields can induce extensive responses in materials. In this study, we report on two kinds of nonlinear mechanisms in the topological semimetal PtSe2 crystal under the excitation of intense terahertz (THz) pulses, which are manipulated by the real and imaginary parts of the nonlinear susceptibility of PtSe2. Regarding the real part, the broken inversion symmetry of PtSe2 is achieved through a THz-electric-field polarization approach, which is characterized by second harmonic generation (SHG) measurements. The transient THz-laser-induced SHG signal occurs within 100 fs and recombines to the equilibrium state within 1 ps, along with a high signal-to-noise ratio (∼51 dB) and a high on/off ratio (∼102). Regarding the imaginary part, a nonlinear absorption change can be generated in the media. We reveal a THz-induced absorption enhancement in PtSe2 via nonlinear transmittance measurements, and the sheet conductivity can be modulated up to 42% by THz electric fields in our experiment. Therefore, the THz-induced ultrafast nonlinear photoresponse reveals the application potential of PtSe2 in photonic and optoelectronic devices in the THz technology.

Inherited KDM6AA649T facilitates tumor-immune escape and exacerbates colorectal signet-ring cell carcinoma outcomes.

Childhood onset of colorectal signet-ring cell carcinoma (CR-SRCC) is extremely rare and featured as highly malignant with poor prognosis. Here we reported a CR-SRCC case of 11-year-old boy with a novel inherited X-linked KDM6AA694T mutation. The H3K27me3 demethylase KDM6A was frequently mutated in varieties of tumors and acts as a tumor suppressor. In vivo H3K27me3 demethylation assay demonstrated that KDM6AA694T had dampened H3K27me3 demethylase activity. Overexpression of KDM6AA694T in SRCC cell line KATO3 promoted cell proliferation, invasion and migration, which were further confirmed in vivo by constructing orthotopic tumor growth and lung metastasis model. Besides, expression of KDM6AA694T in immune cells suppresses inflammatory macrophage response and effector T cell response. In conclusion, we characterized a novel inherited KDM6AA694T mutant from a childhood-onset SRCC case and demonstrated that the mutant with impaired H3K27me3 demethylase activity could potentiate tumor malignancy and suppress antitumor immunity.

Multitask Adversarial Networks Based on Extensive Nonlinear Spiking Neuron Models.

Deep learning technology has been successfully used in Chest X-ray (CXR) images of COVID-19 patients. However, due to the characteristics of COVID-19 pneumonia and X-ray imaging, the deep learning methods still face many challenges, such as lower imaging quality, fewer training samples, complex radiological features and irregular shapes. To address these challenges, this study first introduces an extensive NSNP-like neuron model, and then proposes a multitask adversarial network architecture based on ENSNP-like neurons for chest X-ray images of COVID-19, called MAE-Net. The MAE-Net serves two tasks: (i) converting low-quality CXR images to high-quality images; (ii) classifying CXR images of COVID-19. The adversarial architecture of MAE-Net uses two generators and two discriminators, and two new loss functions have been introduced to guide the optimization of the network. The MAE-Net is tested on four benchmark COVID-19 CXR image datasets and compared them with eight deep learning models. The experimental results show that the proposed MAE-Net can enhance the conversion quality and the accuracy of image classification results.

Study on preparation and properties of silver alloy for Mongolian medicine acupuncture.

The Mongolian medical silver needles often encounter issues of bending, fracturing, and blunting in clinical applications. Similarly, Mongolian warm needles can cause burns on patients due to inaccurate temperature control. In this study, we developed an Ag85Cu15 alloy specifically for acupuncture needles based on material preparation. By incorporating appropriate amounts of Mn and Ti elements, we were able to enhance the mechanical properties and biocompatibility of the acupuncture needles. Compared to commercially available silver needles, this alloy exhibited a significant increase in microhardness up to 210.2 Hv0.2 and an improved tensile strength of 880.2 MPa. Furthermore, we designed a thermoelectric effect-based temperature measurement model for precise control of the warm needle's temperature, enhancing the therapeutic effectiveness of the treatment.

Involution Transformer based U-Net for Landmark Detection in Ultrasound Images for Diagnosis of Infantile DDH.

The B-mode ultrasound based computer-aided diagnosis (CAD) has demonstrated its effectiveness for diagnosis of Developmental Dysplasia of the Hip (DDH) in infants, which can conduct the Graf's method by detecting landmarks in hip ultrasound images. However, it is still necessary to explore more valuable information around these landmarks to enhance feature representation for improving detection performance in the detection model. To this end, a novel Involution Transformer based U-Net (IT-UNet) network is proposed for hip landmark detection. The IT-UNet integrates the efficient involution operation into Transformer to develop an Involution Transformer module (ITM), which consists of an involution attention block and a squeeze-and-excitation involution block. The ITM can capture both the spatial-related information and long-range dependencies from hip ultrasound images to effectively improve feature representation. Moreover, an Involution Downsampling block (IDB) is developed to alleviate the issue of feature loss in the encoder modules, which combines involution and convolution for the purpose of downsampling. The experimental results on two DDH ultrasound datasets indicate that the proposed IT-UNet achieves the best landmark detection performance, indicating its potential applications.

Re-exploration of immunotherapy targeting EMT of hepatocellular carcinoma: Starting from the NF-κB pathway.

Hepatocellular carcinoma (HCC) is the fifth most common malignancies worldwide, and its high morbidity and mortality have brought a heavy burden to the global public health system. Due to the concealment of its onset, the limitation of treatment, the acquisition of multi-drug resistance and radiation resistance, the treatment of HCC cannot achieve satisfactory results. Epithelial mesenchymal transformation (EMT) is a key process that induces progression, distant metastasis, and therapeutic resistance to a variety of malignant tumors, including HCC. Therefore, targeting EMT has become a promising tumor immunotherapy method for HCC. The NF-κB pathway is a key regulatory pathway for EMT. Targeting this pathway has shown potential to inhibit HCC infiltration, invasion, distant metastasis, and therapeutic resistance. At present, there are still some controversies about this pathway and new ideas of combined therapy, which need to be further explored. This article reviews the progress of immunotherapy in improving EMT development in HCC cells by exploring the mechanism of regulating EMT.

Pearl-Inspired Intelligent Marine Hetero Nanocomposite Coating Based on "Brick&Mortar" Strategy: Anti-corrosion Durability and Switchable Antifouling.

Corrosion activities and biofouling pose significant challenges for marine facilities, resulting in substantial economic losses. Inspired by the "brick&mortar" structure of pearls, a novel nanocomposite coating (Pun-HJTx) with long-lasting anti-corrosion and intelligent antifouling modes is fabricated by integrating a compatible MoS2/MXene heterostructure as the "brick" into a polyurea-modified PDMS (Pun) acting as "mortar". Notably, the presence of multiple hydrogen bonds within the coating effectively reduces the pinholes resulted from solution volatilizing. In the dark, where fouling adhesion and microbial corrosion activities are weakened, the MoS2/MXene plays a role in contact bactericidal action. Conversely, during daylight when fouling adhesion and microbial corrosion activities intensify, the coating releases reactive oxygen species (such as hydroxyl radicals and superoxide ions) to counteract fouling adhesion. Additionally, the coating exhibits multi-source self-healing performance under heated or exposed to light (maximum self-healing rate can reach 99.46%) and proves efficient self-cleaning performance and adhesion strength (>2.0 Mpa), making it highly suitable for various practical marine applications. Furthermore, the outstanding performance of the Pun-HJT1 is maintained for approximately 180 days in real-world marine conditions, which proving its practicality and feasibility in real shallow sea environments. This article is protected by copyright. All rights reserved.

Mulberry (Morus alba L.) leaf powder modified the processing of meat alternatives: Principal component analysis from apparent properties to chemical bonds.

For texture control in plant-meat alternatives, the interrelationship between apparent characteristics and chemical bonds in high-fiber formulations remains unclear. The influence of mulberry leaf powder on apparent characteristics and chemical bonds of raw materials, block and strip products at addition amounts of 0.5-25% was analyzed. The results showed that 8% addition significantly increased the chewiness of the block by 98.12%. The strips' texture shows a downward trend, and the processing produced more redness and color difference. Additives promoted the formation of voids, lamellar and filamentous structures, and the strip produced more striped structures. Disulfide bonds significantly increased in the block, and the ß-turn in the secondary structure enhanced by 12.20%. The ß-turn transformed into a ß-sheet in strips. Principal component analysis revealed that the texture improvement was associated with producing disulfide bonds and ß-turn, providing a basis for high-fiber components to improve products' apparent characteristics by chemical bonds.

Co-assisted strategy of sacrificial salt-template and nitrogen-doping to promote lithium storage performance of NiO-Ni/N-C frameworks.

Tailoring the omnidirectional conductivity networks in nickel oxide-based electrodes is important for ensuring their long lifespan, stability, high capacity, and high-rate capability. In this study, nickel metal nanoparticles and a three-dimensional nitrogen-doped carbon matrix were used to embellish the nickel oxide composite NiO-Ni/N-C via simplified hard templating. When a porous nitrogen-doped carbon matrix is present, a rapid pathway would be established for charging and discharging the electrons and lithium ions in a lithium-ion battery, thereby alleviating the volumetric expansion of the NiO nanoparticles during the operation of the battery. Moreover, the Ni0 ions added to serve as active sites to improve the capacity of the NiO-based electrodes and strengthen their conductivities. The multielement-effects of the optimal NiO-Ni/N-C electrode leads it to exhibit a capacity of 1310.8 mAh g-1 at 0.1 A g-1 for 120 loops and a rate capability of 441.5 mAh g-1 at 20.0 A g-1. Kinetic analysis of the prepared electrodes proved their ultrafast ionic and electronic conductivities. This strategy of hard templating reduces the number of routes required for preparing different types of electrodes, including NiO-based electrodes, and improves their electrochemical performance to enable their use in energy storage applications.

Xinyang tablet alleviated cardiac dysfunction in a cardiac pressure overload model by regulating the receptor-interacting serum/three-protein kinase 3/FUN14 domain containing 1-mediated mitochondrial unfolded protein response and mitophagy.

ETHNOPHARMACOLOGICAL RELEVANCE: Xinyang tablet (XYT) has been used for heart failure (HF) for over twenty years in clinical practice, but the underlying molecular mechanism remains poorly understood. AIMS OF THE STUDY: In the present study, we aimed to explore the protective effects of XYT in HF in vivo and in vitro. MATERIALS AND METHODS: Transverse aortic constriction was performed in vivo to establish a mouse model of cardiac pressure overload. Echocardiography, tissue staining, and real-time quantitative PCR (qPCR) were examined to evaluate the protective effects of XYT on cardiac function and structure. Adenosine 5'-triphosphate production, reactive oxygen species staining, and measurement of malondialdehyde and superoxide dismutase was used to detect mitochondrial damage. Mitochondrial ultrastructure was observed by transmission electron microscope. Immunofluorescence staining, qPCR, and Western blotting were performed to evaluate the effect of XYT on the mitochondrial unfolded protein response and mitophagy, and to identify its potential pharmacological mechanism. In vitro, HL-1 cells and neonatal mouse cardiomyocytes were stimulated with Angiotensin II to establish the cell model. Western blotting, qPCR, immunofluorescence staining, and flow cytometry were utilized to determine the effects of XYT on cardiomyocytes. HL-1 cells overexpressing receptor-interacting serum/three-protein kinase 3 (RIPK3) were generated by transfection of RIPK3-overexpressing lentiviral vectors. Cells were then co-treated with XYT to determine the molecular mechanisms. RESULTS: In the present study, XYT was found to exerta protective effect on cardiac function and structure in the pressure overload mice. And it was also found XYT reduced mitochondrial damage by enhancing mitochondrial unfolded protein response and restoring mitophagy. Further studies showed that XYT achieved its cardioprotective role through regulating the RIPK3/FUN14 domain containing 1 (FUNDC1) signaling. Moreover, the overexpression of RIPK3 successfully reversed the XYT-induced protective effects and significantly attenuated the positive effects on the mitochondrial unfolded protein response and mitophagy. CONCLUSIONS: Our findings indicated that XYT prevented pressure overload-induced HF through regulating the RIPK3/FUNDC1-mediated mitochondrial unfolded protein response and mitophagy. The information gained from this study provides a potential strategy for attenuating mitochondrial damage in the context of pressure overload-induced heart failure using XYT.

Advances in DNA-based electrochemical biosensors for the detection of foodborne pathogenic bacteria.

The detection of foodborne pathogenic bacteria is critical in preventing foodborne diseases. DNA-based electrochemical biosensors, with the merits of high sensitivity and short detection time, provide an effective detecting method for foodborne pathogens, attracting significant interest for the past few years. This review mainly describes the important research progress of DNA-based electrochemical biosensors for the detection of foodborne pathogenic bacteria through four perspectives: representative foodborne pathogens detection using electrochemical approaches, DNA immobilization strategies of aptamers, DNA-based signal amplification strategies used in electrochemical DNA sensors, and functional DNA used in electrochemical DNA sensors. Finally, perspectives and challenges are presented in this field. This review will contribute to DNA-based electrochemical biosensor in enhancing the nucleic acid signal amplification.

Is postmastectomy radiotherapy necessary for breast cancer patients with clinically node-positive downstaging to ypN0 after neoadjuvant chemotherapy?

PURPOSE: The significance of postmastectomy radiotherapy (PMRT) in breast cancer patients who initially have clinically node-positive (cN +) status but achieve downstaging to ypN0 following neoadjuvant chemotherapy (NAC) remains uncertain. This study aims to assess the impact of PMRT in this patient subset. METHODS: Patients were enrolled from West China Hospital, Sichuan University from 2008 to 2019. Overall survival (OS), Locoregional recurrence-free survival (LRFS), distant metastasis-free survival (DMFS), and breast cancer-specific survival (BCSS) were estimated using the Kaplan-Meier method and assessed with the log-rank test. The impact of PMRT was further analyzed by the Cox proportional hazards model. Propensity score matching (PSM) was performed to reduce the selection bias. RESULTS: Of the 333 eligible patients, 189 (56.8%) received PMRT, and 144 (43.2%) did not. At a median follow-up period of 71 months, the five-year LRFS, DMFS, BCSS, and OS rates were 99.1%, 93.4%, 96.4%, and 94.3% for the entire cohort, respectively. Additionally, the 5-year LRFS, DMFS, BCSS, and OS rates were 98.9%, 93.8%, 96.7%, and 94.5% with PMRT and 99.2%, 91.3%, 94.9%, and 92.0% without PMRT, respectively (all p-values not statistically significant). After multivariate analysis, PMRT was not a significant risk factor for any of the endpoints. When further stratified by stage, PMRT did not show any survival benefit for patients with stage II-III diseases. CONCLUSION: In the context of comprehensive treatments, PMRT might be exempted in ypN0 breast cancer patients. Further large-scale, randomized controlled studies are required to investigate the significance of PMRT in this patient subset.

Multiple-in-Single-Out Object Detector Leveraging Spiking Neural Membrane Systems and Multiple Transformers.

Most existing multi-scale object detectors depend on multi-level feature maps. The Feature Pyramid Networks (FPN) is a significant architecture for object detection that utilizes these multi-level feature maps. However, the use of FPN also increases the detector's complexity. For object detection methods that only use a single-level feature map, the detection performance is limited to some extent because the single-level feature map cannot balance deep semantic information and shallow detail information. We introduce a novel detector - the Spiking Neural P Multiple-in-Single-out (SNPMiSo) detector to address these challenges. The SNPMiSo detector is constructed based on SNP-like neurons. In SNPMiSo, we employ two kinds of Transformers to boost the important features across different-level feature maps separately. After enhancing the features, we use an incremental upsampling module to upsample and merge the two feature maps. This combined feature map is input into the NAF dilated residual module and the NAF dual-branch detection head. This process allows us to extract multi-scale features and carry out detection tasks. Our tests show promising results: On the COCO dataset, SNPMiSo attains an Average Precision (AP) of 38.7, an improvement of 1.0 AP over YOLOF. In addition, SNPMiSo demonstrates a quicker detection speed, outperforming some advanced multi-level and single-level object detectors.

Distal Protein-Protein Interactions Contribute to SARS-CoV-2 Main Protease Substrate Binding and Nirmatrelvir Resistance.

SARS-CoV-2 main protease, M pro , is responsible for the processing of the viral polyproteins into individual proteins, including the protease itself. M pro is a key target of anti-COVID-19 therapeutics such as nirmatrelvir (the active component of Paxlovid). Resistance mutants identified clinically and in viral passage assays contain a combination of active site mutations (e.g. E166V, E166A, L167F), which reduce inhibitor binding and enzymatic activity, and non-active site mutations (e.g. P252L, T21I, L50F), which restore the fitness of viral replication. Although the mechanism of resistance for the active site mutations is apparent, the role of the non-active site mutations in fitness rescue remains elusive. In this study, we use the model system of a M pro triple mutant (L50F/E166A/L167F) that confers not only nirmatrelvir drug resistance but also a similar fitness of replication compared to the wild-type both in vitro and in vivo. By comparing peptide and full-length M pro protein as substrates, we demonstrate that the binding of M pro substrate involves more than residues in the active site. In particular, L50F and other non-active site mutations can enhance the M pro dimer-dimer interactions and help place the nsp5-6 substrate at the enzyme catalytic center. The structural and enzymatic activity data of M pro L50F, L50F/E166A/L167F, and others underscore the importance of considering the whole substrate protein in studying M pro and substrate interactions, and offers important insights into M pro function, resistance development, and inhibitor design.

Cost-Sensitive Weighted Contrastive Learning Based on Graph Convolutional Networks for Imbalanced Alzheimer's Disease Staging.

Identifying the progression stages of Alzheimer's disease (AD) can be considered as an imbalanced multi-class classification problem in machine learning. It is challenging due to the class imbalance issue and the heterogeneity of the disease. Recently, graph convolutional networks (GCNs) have been successfully applied in AD classification. However, these works did not handle the class imbalance issue in classification. Besides, they ignore the heterogeneity of the disease. To this end, we propose a novel cost-sensitive weighted contrastive learning method based on graph convolutional networks (CSWCL-GCNs) for imbalanced AD staging using resting-state functional magnetic resonance imaging (rs-fMRI). The proposed method is developed on a multi-view graph constructed using the functional connectivity (FC) and high-order functional connectivity (HOFC) features of the subjects. A novel cost-sensitive weighted contrastive learning procedure is proposed to capture discriminative information from the minority classes, encouraging the samples in the minority class to provide adequate supervision. Considering the heterogeneity of the disease, the weights of the negative pairs are introduced into contrastive learning and they are computed based on the distance to class prototypes, which are automatically learned from the training data. Meanwhile, the cost-sensitive mechanism is further introduced into contrastive learning to handle the class imbalance issue. The proposed CSWCL-GCN is evaluated on 720 subjects (including 184 NCs, 40 SMC patients, 208 EMCI patients, 172 LMCI patients and 116 AD patients) from the ADNI (Alzheimer's Disease Neuroimaging Initiative). Experimental results show that the proposed CSWCL-GCN outperforms state-of-the-art methods on the ADNI database.