PED: a novel predictor-encoder-decoder model for Alzheimer drug molecular generation.

Alzheimer's disease (AD) is a gradually advancing neurodegenerative disorder characterized by a concealed onset. Acetylcholinesterase (AChE) is an efficient hydrolase that catalyzes the hydrolysis of acetylcholine (ACh), which regulates the concentration of ACh at synapses and then terminates ACh-mediated neurotransmission. There are inhibitors to inhibit the activity of AChE currently, but its side effects are inevitable. In various application fields where Al have gained prominence, neural network-based models for molecular design have recently emerged and demonstrate encouraging outcomes. However, in the conditional molecular generation task, most of the current generation models need additional optimization algorithms to generate molecules with intended properties which make molecular generation inefficient. Consequently, we introduce a cognitive-conditional molecular design model, termed PED, which leverages the variational auto-encoder. Its primary function is to adeptly produce a molecular library tailored for specific properties. From this library, we can then identify molecules that inhibit AChE activity without adverse effects. These molecules serve as lead compounds, hastening AD treatment and concurrently enhancing the AI's cognitive abilities. In this study, we aim to fine-tune a VAE model pre-trained on the ZINC database using active compounds of AChE collected from Binding DB. Different from other molecular generation models, the PED can simultaneously perform both property prediction and molecule generation, consequently, it can generate molecules with intended properties without additional optimization process. Experiments of evaluation show that proposed model performs better than other methods benchmarked on the same data sets. The results indicated that the model learns a good representation of potential chemical space, it can well generate molecules with intended properties. Extensive experiments on benchmark datasets confirmed PED's efficiency and efficacy. Furthermore, we also verified the binding ability of molecules to AChE through molecular docking. The results showed that our molecular generation system for AD shows excellent cognitive capacities, the molecules within the molecular library could bind well to AChE and inhibit its activity, thus preventing the hydrolysis of ACh.

Heterogeneous graph inference with range constrainted L2,1-collaborative matrix factorization for small molecule-miRNA association prediction.

MicroRNAs (miRNAs) play a vital role in regulating gene expression and various biological processes. As a result, they have been identified as effective targets for small molecule (SM) drugs in disease treatment. Heterogeneous graph inference stands as a classical approach for predicting SM-miRNA associations, showcasing commendable convergence accuracy and speed. However, most existing methods do not adequately address the inherent sparsity in SM-miRNA association networks, and imprecise SM/miRNA similarity metrics reduce the accuracy of predicting SM-miRNA associations. In this research, we proposed a heterogeneous graph inference with range constrained L2,1-collaborative matrix factorization (HGIRCLMF) method to predict potential SM-miRNA associations. First, we computed the multi-source similarities of SM/miRNA and integrated these similarity information into a comprehensive SM/miRNA similarity. This step improved the accuracy of SM and miRNA similarity, ensuring reliability for the subsequent inference of the heterogeneity map. Second, we used a range constrained L2,1-collaborative matrix factorization (RCLMF) model to pre-populate the SM-miRNA association matrix with missing values. In this step, we developed a novel matrix decomposition method that enhances the robustness and formative nature of SM-miRNA edges between SM networks and miRNA networks. Next, we built a well-established SM-miRNA heterogeneous network utilizing the processed biological information. Finally, HGIRCLMF used this network data to infer unknown association pair scores. We implemented four cross-validation experiments on two distinct datasets, and HGIRCLMF acquired the highest areas under the curve, surpassing six state-of-the-art computational approaches. Furthermore, we performed three case studies to validate the predictive power of our method in practical application.

A new method for accurate inversion of Forel-Ule index using MODIS images - revealing the water color evolution in China's large lakes and reservoirs over the past two decades.

The Forel Ule water color index (FUI) based on satellite inversion can characterize the comprehensive characteristics of water quality on a large spatiotemporal scale. The high-frequency observations and rich historical data of the MODIS surface reflectance product (MODIS-500 m) provide important data support for monitoring the FUI of inland lakes. However, MODIS-500 m has only three bands in the visible light range, resulting in significant uncertainty in FUI inversion. To address this problem, this study developed an improved FUI inversion model using 500 synthetic spectra covering natural waters. The model, with a performance threshold set at 170° (FUI = 8), used a segmented algorithm across the entire color space. Validated with on-site measurement datasets (3500 samples), the model exhibited excellent performance, with mean relative error (MRE) and root mean square error (RMSE) of 1.71 % and 3.63°, respectively. Compared to existing models, it was more suitable for long-term FUI inversion in various types of lakes, particularly in eutrophic regions. Subsequently, the model was applied to MODIS-500 m observations from 2000 to 2022, revealing the spatiotemporal dynamics of FUI in 180 large lakes and reservoirs (hereinafter referred to as lakes) in China. The results indicated that the long-term mean FUI in the study area was 9, with 7 and 12 in the western and eastern regions, respectively, showing a distinct spatial distribution of "blue in the west and green in the east." The mean change rate of hue angle for all lakes was -0.085°/yr, showing an overall decreasing trend. Environmental factors' relative contributions to long-term water color changes in each lake region were quantified using the multiple general linear model (GLM). Although each lake region exhibited different driving forces, they were primarily influenced by vegetation, lake surface area, and anthropogenic factors. Additionally, the seasonal types of lake water color were analyzed, with the west and east showing opposite patterns, reflecting the significant influence of topographic features and seasonal changes in climate on water color. The research results provide techniques for accurate inversion of FUI using MODIS-500 m data, while deepening the understanding of long-term water color changes in inland lakes in China.

CHL-DTI: A Novel High-Low Order Information Convergence Framework for Effective Drug-Target Interaction Prediction.

Recognizing drug-target interactions (DTI) stands as a pivotal element in the expansive field of drug discovery. Traditional biological wet experiments, although valuable, are time-consuming and costly as methods. Recently, computational methods grounded in network learning have demonstrated great advantages by effective topological feature extraction and attracted extensive research attention. However, most existing network-based learning methods only consider the low-order binary correlation between individual drug and target, neglecting the potential higher-order correlation information derived from multiple drugs and targets. High-order information, as an essential component, exhibits complementarity with low-order information. Hence, the incorporation of higher-order associations between drugs and targets, while adequately integrating them with the existing lower-order information, could potentially yield substantial breakthroughs in predicting drug-target interactions. We propose a novel dual channels network-based learning model CHL-DTI that converges high-order information from hypergraphs and low-order information from ordinary graph for drug-target interaction prediction. The convergence of high-low order information in CHL-DTI is manifested in two key aspects. First, during the feature extraction stage, the model integrates both high-level semantic information and low-level topological information by combining hypergraphs and ordinary graph. Second, CHL-DTI fully fuse the innovative introduced drug-protein pairs (DPP) hypergraph network structure with ordinary topological network structure information. Extensive experimentation conducted on three public datasets showcases the superior performance of CHL-DTI in DTI prediction tasks when compared to SOTA methods. The source code of CHL-DTI is available at https://github.com/UPCLyy/CHL-DTI .

Investigation of the molecular mechanism of Danggui Buxue tang in treating lung cancer using network pharmacology and molecular docking techniques.

This study used network pharmacology and molecular docking techniques to investigate the molecular targets and pathways of Danggui Buxue Tang (DBT) in treating lung cancer. The compound-target network was constructed using the Traditional Chinese Medicine Systems Pharmacology Database (TCMSP), and a lung cancer-specific network was created using the GEO database and Cytoscape software. GO and KEGG pathway analyses were performed to understand the biological processes associated with DBT. The key compounds from Astragalus, kaempferol, and quercetin, and the potential targets are IL-6, IL-1ß, FOS, ICAM1, and CCL2. GO enrichment analysis revealed numerous biological process-related entries, while KEGG pathway analysis highlighted the TNF and IL-17 signalling pathways. Molecular docking confirmed the stable binding activity between the main active compounds of DBT and the target proteins. Overall, these findings shed light on the molecular mechanism of DBT in treating lung cancer, providing insights into targets, pathways, and biological processes involved.

A comprehensive review of the literature on CD10: its function, clinical application, and prospects.

CD10, a zinc-dependent metalloprotease found on the cell surface, plays a pivotal role in an array of physiological and pathological processes including cardiovascular regulation, immune function, fetal development, pain response, oncogenesis, and aging. Recognized as a biomarker for hematopoietic and tissue stem cells, CD10 has garnered attention for its prognostic potential in the progression of leukemia and various solid tumors. Recent studies underscore its regulatory significance and therapeutic promise in combating Alzheimer's disease (AD), and it is noted for its protective role in preventing heart failure (HF), obesity, and type-2 diabetes. Furthermore, CD10/substance P interaction has also been shown to contribute to the pain signaling regulation and immunomodulation in diseases such as complex regional pain syndrome (CRPS) and osteoarthritis (OA). The emergence of COVID-19 has sparked interest in CD10's involvement in the disease's pathogenesis. Given its association with multiple disease states, CD10 is a prime therapeutic target; inhibitors targeting CD10 are now being advanced as therapeutic agents. This review compiles recent and earlier literature on CD10, elucidating its physicochemical attributes, tissue-specific expression, and molecular functions. Furthermore, it details the association of CD10 with various diseases and the clinical advancements of its inhibitors, providing a comprehensive overview of its growing significance in medical research.

scSID: A lightweight algorithm for identifying rare cell types by capturing differential expression from single-cell sequencing data.

Single-cell RNA sequencing (scRNA-seq) is currently an important technology for identifying cell types and studying diseases at the genetic level. Identifying rare cell types is biologically important as one of the downstream data analyses of single-cell RNA sequencing. Although rare cell identification methods have been developed, most of these suffer from insufficient mining of intercellular similarities, low scalability, and being time-consuming. In this paper, we propose a single-cell similarity division algorithm (scSID) for identifying rare cells. It takes cell-to-cell similarity into consideration by analyzing both inter-cluster and intra-cluster similarities, and discovers rare cell types based on the similarity differences. We show that scSID outperforms other existing methods by benchmarking it on different experimental datasets. Application of scSID to multiple datasets, including 68K PBMC and intestine, highlights its exceptional scalability and remarkable ability to identify rare cell populations.

Selective inhibition of CDK9 in triple negative breast cancer.

Targeted therapy for triple-negative breast cancers (TNBC) remains a clinical challenge due to tumour heterogeneity. Since TNBC have key features of transcriptionally addicted cancers, targeting transcription via regulators such as cyclin-dependent kinase 9 (CDK9) has potential as a therapeutic strategy. Herein, we preclinically tested a new selective CDK9 inhibitor (CDDD11-8) in TNBC using cell line, patient-derived organoid, and patient-derived explant models. In vitro, CDDD11-8 dose-dependently inhibited proliferation (IC50 range: 281-734 nM), induced cell cycle arrest, and increased apoptosis of cell lines, which encompassed the three major molecular subtypes of TNBC. On target inhibition of CDK9 activity was demonstrated by reduced RNAPII phosphorylation at a CDK9 target peptide and down-regulation of the MYC and MCL1 oncogenes at the mRNA and protein levels in all cell line models. Drug induced RNAPII pausing was evident at gene promoters, with strongest pausing at MYC target genes. Growth of five distinct patient-derived organoid models was dose-dependently inhibited by CDDD11-8 (IC50 range: 272-771 nM), including three derived from MYC amplified, chemo-resistant TNBC metastatic lesions. Orally administered CDDD11-8 also inhibited growth of mammary intraductal TNBC xenograft tumours with no overt toxicity in vivo (mice) or ex vivo (human breast tissues). In conclusion, our studies indicate that CDK9 is a viable therapeutic target in TNBC and that CDDD11-8, a novel selective CDK9 inhibitor, has efficacy in TNBC without apparent toxicity to normal tissues.

SESN2-Mediated AKT/GSK-3ß/NRF2 Activation to Ameliorate Adriamycin Cardiotoxicity in High-Fat Diet-Induced Obese Mice.

Aims: Obese patients are highly sensitive to adriamycin (ADR)-induced cardiotoxicity. However, the potential mechanism of superimposed toxicity remains to be elucidated. Sestrin 2 (SESN2), a potential antioxidant, could attenuate stress-induced cardiomyopathy; therefore, this study aims to explore whether SESN2 enhances cardiac resistance to ADR-induced oxidative damage in high-fat diet (HFD)-induced obese mice. Results: The results revealed that obesity decreased SESN2 expression in ADR-exposed heart. And, HFD mice may predispose to ADR-induced cardiotoxicity, which was probably associated with inhibiting protein kinase B (AKT), glycogen synthase kinase-3 beta (GSK-3ß) phosphorylation and subsequently blocking nuclear localization of nuclear factor erythroid-2 related factor 2 (NRF2), ultimately resulting in cardiac oxidative damage. However, these destructive cascades and cardiac oxidative damage effects induced by HFD/sodium palmitate combined with ADR were blocked by overexpression of SESN2. Moreover, the antioxidant effect of SESN2 could be largely abolished by sh-Nrf2 or wortmannin. And sulforaphane, an NRF2 agonist, could remarkably reverse cardiac pathological and functional abnormalities caused by ADR in obese mice. Innovation and Conclusion: This study demonstrated that SESN2 might be a promising therapeutic target for improving anthracycline-related cardiotoxicity in obesity by upregulating activity of NRF2 via AKT/GSK-3ß/Src family tyrosine kinase signaling pathway. Antioxid. Redox Signal. 40, 598-615.

Correlation between visceral fat metabolism score and erectile dysfunction: a cross-sectional study from NHANES 2001-2004.

Backgrounds: The factors associated with erectile dysfunction (ED) are diverse, and obesity is a significant component. Metabolic Score for Visceral Fat (METS-VF) can assess obesity more accurately than body mass index (BMI). However, the association between METS-VF and ED remains unclear. Objective: This study aimed to investigate the association between the METS-VF and ED using National Health and Nutrition Examination Survey (NHANES) 2001-2004 data. Methods: Data were sourced from NHANES 2001-2004. The relationship between METS-VF and ED was analyzed using multivariate logistic regression, followed by subgroup analyses to identify sensitive populations. Nonlinear correlation was evaluated through smoothed curve fitting, and a threshold effect analysis validated the findings. Comparative logistic regression of the Receiver Operating Characteristic (ROC) curve assessed the diagnostic capability of METS-VF against the classical obesity index for ED. Results: The study enrolled 3625 participants, of whom 961 self-reported ED history and 360 reported severe ED. After adjusting for confounders, METS-VF exhibited a positive association with asthma prevalence (OR= 3.47, 95% CI: 2.83, 14.24). Stratification based on median METS-VF revealed higher ED prevalence in participants with elevated METS-VF (OR= 2.81,95% CI:2.32, 3.41). Nonlinear correlation was observed, with a significant association between METS-VF and ED when METS-VF exceeded 6.63. Subgroup analysis highlighted a stronger correlation in participants aged 50-85 years, Caucasians, hypertensive individuals, diabetics, and those with coronary heart disease. Sensitivity analysis using severe ED as the outcome reaffirmed the nonlinear positive association with METS-VF (OR=3.86, 95% CI:2.80,5.33), particularly when METS-VF surpassed 6.68. Conclusion: Elevated METS-VF was nonlinearly correlated with increased ED incidence. Individuals with METS-VF above 6.63 should be vigilant about heightened ED risk. Special attention should be given to participants aged 50-85 years, Caucasians, hypertensive individuals, diabetics, and those with coronary heart disease.

Basic research of meridian-tendon based on fascia: review and prospects. / åŸºäºŽç­‹è†œçš„ç»ç­‹åŸºç¡€ç ”ç©¶ï¼šè¿°è¯„ä¸Žå±•æœ›.

Meridian-tendon is a central concept in meridian theory of TCM, and its basic research has been increasingly emphasized. While there is no unified understanding of the essence of meridian-tendon, the concept that function of fascia could partially reflect the functions of meridian-tendons has reached consensus in the academic community. This article suggests that under the guidance of meridian-tendon theory, based on previous research foundation of fascia, focusing on adopting fascia research methods, the mechanisms of tender point hyperalgesia and abnormal proliferation related to meridian lesions should be adopted to explain yitong weishu (taking the worst painful sites of muscle spasm as the points), and the mechanisms of meridian intervention efficacy should be adopted to explain yizhi weishu (feelings from patients and acupuncture operators). Furthermore, this article provides an analysis of the future trends in basic research of meridian tendons.

Identifying potential small molecule-miRNA associations via Robust PCA based on γ-norm regularization.

Dysregulation of microRNAs (miRNAs) is closely associated with refractory human diseases, and the identification of potential associations between small molecule (SM) drugs and miRNAs can provide valuable insights for clinical treatment. Existing computational techniques for inferring potential associations suffer from limitations in terms of accuracy and efficiency. To address these challenges, we devise a novel predictive model called RPCA$\Gamma $NR, in which we propose a new Robust principal component analysis (PCA) framework based on $\gamma $-norm and $l_{2,1}$-norm regularization and design an Augmented Lagrange Multiplier method to optimize it, thereby deriving the association scores. The Gaussian Interaction Profile Kernel Similarity is calculated to capture the similarity information of SMs and miRNAs in known associations. Through extensive evaluation, including Cross Validation Experiments, Independent Validation Experiment, Efficiency Analysis, Ablation Experiment, Matrix Sparsity Analysis, and Case Studies, RPCA$\Gamma $NR outperforms state-of-the-art models concerning accuracy, efficiency and robustness. In conclusion, RPCA$\Gamma $NR can significantly streamline the process of determining SM-miRNA associations, thus contributing to advancements in drug development and disease treatment.

[Construction of jingjin differentiation system based on holistic concept].

As a diagnostic method to guide the treatment of sinew/fascia diseases, jingjin (muscle regions of meridians) differentiation is an important component of syndrome differentiation system of acupuncture and moxibustion. In clinical practice, because of the limitations of the ideological guidance of the holistic view, the systemic and dialectical thinking and the syndrome element collection, the system of diagnosis and treatment of sinew/fascia diseases is not comprehensive. Through combing the origin of the holistic view of jingjin, the paper expounds the differentiation framework of sinew/fascia diseases from 4 aspects of differentiation, i.e. the location of disease, etiology, nature of disease and condition of disease. It suggests to construct jingjin differentiation system by taking the holistic ideas as the core, the syndrome element research as the common method and the evidence-based medicine as the theoretical basis so that the thinking of syndrome differentiation and the diagnostic approaches based on jingjin theory can be enriched.

Enhancements on volatile organic compounds (VOCs) adsorption and desorption performance of ZSM-5 by fabricating hierarchical MCM-41.

ZSM-5 zeolite has been considered a promising adsorbent for capturing VOCs. However, its hydrophilicity and narrow micropore structure limit their practical application especially under humid atmospheres. In this study, the pure silica mesoporous molecular sieve MCM-41 was assembled on ZSM-5 zeolite with different SiO2/Al2O3 ratios (SARs) via a surfactant-mediated recrystallization method. Then, its adsorption-desorption behaviors were investigated using n-hexane, toluene, and ethyl acetate as VOC model molecules. The results showed that the hydrophobicity of ZSM-5/MCM-41 composites and their VOC diffusion behaviors were significantly improved. Furthermore, the SARs of the ZSM-5 precursors had a remarkable influence on the adsorption performance of ZSM-5/MCM-41 composites. ZSM-5/MCM-41(130) was the optimum option, and its dynamic adsorption capacity for ethyl acetate (111.30 mg·g-1) was higher than that of the corresponding ZSM-5 zeolites even under statured humidity. Meanwhile, the ratios of dynamic adsorption capacities at humid and dry atmospheres (qs,wet/qs,dry) of ZSM-5/MCM-41(130) for n-hexane, toluene, and ethyl acetate were 84.89%, 61.46%, and 73.81% respectively. The results will provide guidelines for the preparation of hydrophobic adsorbents.

VGAEDTI: drug-target interaction prediction based on variational inference and graph autoencoder.

MOTIVATION: Accurate identification of Drug-Target Interactions (DTIs) plays a crucial role in many stages of drug development and drug repurposing. (i) Traditional methods do not consider the use of multi-source data and do not consider the complex relationship between data sources. (ii) How to better mine the hidden features of drug and target space from high-dimensional data, and better solve the accuracy and robustness of the model. RESULTS: To solve the above problems, a novel prediction model named VGAEDTI is proposed in this paper. We constructed a heterogeneous network with multiple sources of information using multiple types of drug and target dataIn order to obtain deeper features of drugs and targets, we use two different autoencoders. One is variational graph autoencoder (VGAE) which is used to infer feature representations from drug and target spaces. The second is graph autoencoder (GAE) propagating labels between known DTIs. Experimental results on two public datasets show that the prediction accuracy of VGAEDTI is better than that of six DTIs prediction methods. These results indicate that model can predict new DTIs and provide an effective tool for accelerating drug development and repurposing.

Generative Adversarial Matrix Completion Network based on Multi-Source Data Fusion for miRNA-Disease Associations Prediction.

Numerous biological studies have shown that considering disease-associated micro RNAs (miRNAs) as potential biomarkers or therapeutic targets offers new avenues for the diagnosis of complex diseases. Computational methods have gradually been introduced to reveal disease-related miRNAs. Considering that previous models have not fused sufficiently diverse similarities, that their inappropriate fusion methods may lead to poor quality of the comprehensive similarity network and that their results are often limited by insufficiently known associations, we propose a computational model called Generative Adversarial Matrix Completion Network based on Multi-source Data Fusion (GAMCNMDF) for miRNA-disease association prediction. We create a diverse network connecting miRNAs and diseases, which is then represented using a matrix. The main task of GAMCNMDF is to complete the matrix and obtain the predicted results. The main innovations of GAMCNMDF are reflected in two aspects: GAMCNMDF integrates diverse data sources and employs a nonlinear fusion approach to update the similarity networks of miRNAs and diseases. Also, some additional information is provided to GAMCNMDF in the form of a 'hint' so that GAMCNMDF can work successfully even when complete data are not available. Compared with other methods, the outcomes of 10-fold cross-validation on two distinct databases validate the superior performance of GAMCNMDF with statistically significant results. It is worth mentioning that we apply GAMCNMDF in the identification of underlying small molecule-related miRNAs, yielding outstanding performance results in this specific domain. In addition, two case studies about two important neoplasms show that GAMCNMDF is a promising prediction method.

DETHACDA: A Dual-view Edge and Topology Hybrid Attention Model for CircRNA-Disease Associations Prediction.

There exists growing evidence that circRNAs are concerned with many complex diseases physiological processes and pathogenesis and may serve as critical therapeutic targets. Identifying disease-associated circRNAs through biological experiments is time-consuming, and designing an intelligent, precise calculation model is essential. Recently, many models based on graph technology have been proposed to predict circRNA-disease association. However, most existing methods only capture the neighborhood topology of the association network and ignore the complex semantic information. Therefore, we propose a Dual-view Edge and Topology Hybrid Attention model for predicting CircRNA-Disease Associations (DETHACDA), effectively capturing the neighborhood topology and various semantics of circRNA and disease nodes in a heterogeneous network. The 5-fold cross-validation experiments on circRNADisease indicate that the proposed DETHACDA achieves the area under receiver operating characteristic curve of 0.9882, better than four state-of-the-art calculation methods.

Relationship between carbon pool changes and environmental changes in arid and semi-arid steppe-A two decades study in Inner Mongolia, China.

As the arid and semi-arid grassland with the most extensive distribution area in northern China, the carbon stored in Inner Mongolia (IM) grassland is highly susceptible to environmental changes. With the global warming and drastic climate changes, exploring the relationship between carbon pool changes and environmental changes and their spatiotemporal heterogeneity is necessary. This study estimates the carbon pool distribution of IM grassland during 2003-2020 by combining the measured below ground biomass (BGB) dataset, measured soil organic carbon (SOC) dataset, multi-source satellite remote sensing data products, and random forest regression modeling method. It also discusses the variation trend of BGB/SOC and its correlation with critical environmental factors, vegetation condition factors and drought index. The results show that the BGB/SOC in IM grassland was stable during 2003-2020, with a weak upward trend. The correlation analysis reveals that high temperature and drought environment were unfavorable for developing vegetation roots and would lead to a decrease in BGB. Furthermore, temperature rise, soil moisture decrease, and drought adversely effected grassland biomass and SOC in areas with low altitude, high SOC density, suitable temperature and humidity. However, in areas with relatively poor natural environments and relatively low SOC content, SOC was not significantly affected by environmental deterioration and even showed an accumulation trend. These conclusions provide directions for SOC treatment and protection. In areas where SOC is abundant, it is important to reduce carbon loss caused by environmental changes. However, in areas with poor SOC, due to the high carbon storage potential of grasslands, carbon storage can be improved through scientifically managing grazing and protecting vulnerable grasslands.

WRANet: wavelet integrated residual attention U-Net network for medical image segmentation.

Medical image segmentation is crucial for the diagnosis and analysis of disease. Deep convolutional neural network methods have achieved great success in medical image segmentation. However, they are highly susceptible to noise interference during the propagation of the network, where weak noise can dramatically alter the network output. As the network deepens, it can face problems such as gradient explosion and vanishing. To improve the robustness and segmentation performance of the network, we propose a wavelet residual attention network (WRANet) for medical image segmentation. We replace the standard downsampling modules (e.g., maximum pooling and average pooling) in CNNs with discrete wavelet transform, decompose the features into low- and high-frequency components, and remove the high-frequency components to eliminate noise. At the same time, the problem of feature loss can be effectively addressed by introducing an attention mechanism. The combined experimental results show that our method can effectively perform aneurysm segmentation, achieving a Dice score of 78.99%, an IoU score of 68.96%, a precision of 85.21%, and a sensitivity score of 80.98%. In polyp segmentation, a Dice score of 88.89%, an IoU score of 81.74%, a precision rate of 91.32%, and a sensitivity score of 91.07% were achieved. Furthermore, our comparison with state-of-the-art techniques demonstrates the competitiveness of the WRANet network.

scASGC: An adaptive simplified graph convolution model for clustering single-cell RNA-seq data.

Single-cell RNA sequencing (scRNA-seq) is now a successful technique for identifying cellular heterogeneity, revealing novel cell subpopulations, and forecasting developmental trajectories. A crucial component of the processing of scRNA-seq data is the precise identification of cell subpopulations. Although many unsupervised clustering methods have been developed to cluster cell subpopulations, the performance of these methods is vulnerable to dropouts and high dimensionality. In addition, most existing methods are time-consuming and fail to adequately account for potential associations between cells. In the manuscript, we present an unsupervised clustering method based on an adaptive simplified graph convolution model called scASGC. The proposed method builds plausible cell graphs, aggregates neighbor information using a simplified graph convolution model, and adaptively determines the most optimal number of convolution layers for various graphs. Experiments on 12 public datasets show that scASGC outperforms both classical and state-of-the-art clustering methods. In addition, in a study of mouse intestinal muscle containing 15,983 cells, we identified distinct marker genes based on the clustering results of scASGC. The source code of scASGC is available at https://github.com/ZzzOctopus/scASGC.