Construction and Validation of a Rat Model of Acute Necrotizing Pancreatitis-Associated Intestinal Injury.

Acute pancreatitis (AP) is an acute inflammatory reaction of the pancreatic tissue, which involves auto-digestion, oedema, haemorrhage, and necrosis. AP can be categorized into mild, moderately severe and severe AP, with severe pancreatitis also referred to as acute necrotizing pancreatitis (ANP). ANP is characterized by the accumulation of necrotic material in the peritoneal cavity. This can result in intestinal injury. However, the mechanism of ANP-associated intestinal injury remains unclear. We established an ANP-associated intestinal injury rat model (ANP-IR model) by injecting pancreatitis-associated ascites fluid (PAAF) and necrotic pancreatic tissue at various proportions into the triangular area formed by the left renal artery and ureter. The feasibility of the ANP-IR model was verified by comparing the similar changes in indicators of intestinal inflammation and barrier function between the two rat models. In addition, we detected changes in apoptosis levels and YAP protein expression in the ileal tissues of rats in each group and validated them in vitro in rat epithelial crypt cells (IEC-6) to further explore the potential injury mechanisms of ANP-associated intestinal injury. We also collected clinical data from patients with ANP to validate the effects of PAAF and pancreatic necrosis on intestinal injury. Our findings offer a theoretical basis for restricting the buildup of peritoneal necrosis in individuals with ANP, thus promoting the restoration of intestinal function and enhancing treatment efficacy. The use of the ANP-IR model in further studies can help us better understand the mechanism and treatment of ANP-associated intestinal injury.

Research Progress on Taxus Extraction and Formulation Preparation Technologies.

Taxus, as a globally prevalent evergreen tree, contains a wealth of bioactive components that play a crucial role in the pharmaceutical field. Taxus extracts, defined as a collection of one or more bioactive compounds extracted from the genus Taxus spp., have become a significant focus of modern cancer treatment research. This review article aims to delve into the scientific background of Taxus extracts and their considerable value in pharmaceutical research. It meticulously sifts through and compares various advanced extraction techniques such as supercritical extraction, ultrasound extraction, microwave-assisted extraction, solid-phase extraction, high-pressure pulsed electric field extraction, and enzymatic extraction, assessing each technology's advantages and limitations across dimensions such as extraction efficiency, extraction purity, economic cost, operational time, and environmental impact, with comprehensive analysis results presented in table form. In the area of drug formulation design, this paper systematically discusses the development strategies for solid, liquid, and semi-solid dosage forms based on the unique physicochemical properties of Taxus extracts, their intended medical uses, and specific release characteristics, delving deeply into the selection of excipients and the critical technical issues in the drug preparation process. Moreover, the article looks forward to the potential directions of Taxus extracts in future research and medical applications, emphasizing the urgency and importance of continuously optimizing extraction methods and formulation design to enhance treatment efficacy, reduce production costs, and decrease environmental burdens. It provides a comprehensive set of preparation techniques and formulation optimization schemes for researchers in cancer treatment and other medical fields, promoting the application and development of Taxus extracts in pharmaceutical sciences.

Silver Nanoparticles Encapped by Dihydromyricetin: Optimization of Green Synthesis, Characterization, Toxicity, and Anti-MRSA Infection Activities for Zebrafish (Danio rerio).

To achieve the environmentally friendly and rapid green synthesis of efficient and stable AgNPs for drug-resistant bacterial infection, this study optimized the green synthesis process of silver nanoparticles (AgNPs) using Dihydromyricetin (DMY). Then, we assessed the impact of AgNPs on zebrafish embryo development, as well as their therapeutic efficacy on zebrafish infected with Methicillin-resistant Staphylococcus aureus (MRSA). Transmission electron microscopy (TEM) and dynamic light-scattering (DLS) analyses revealed that AgNPs possessed an average size of 23.6 nm, a polymer dispersity index (PDI) of 0.197 ± 0.0196, and a zeta potential of -18.1 ± 1.18 mV. Compared to other published green synthesis products, the optimized DMY-AgNPs exhibited smaller sizes, narrower size distributions, and enhanced stability. Furthermore, the minimum concentration of DMY-AgNPs required to affect zebrafish hatching and survival was determined to be 25.0 µg/mL, indicating the low toxicity of DMY-AgNPs. Following a 5-day feeding regimen with DMY-AgNP-containing food, significant improvements were observed in the recovery of the gills, intestines, and livers in MRSA-infected zebrafish. These results suggested that optimized DMY-AgNPs hold promise for application in aquacultures and offer potential for further clinical use against drug-resistant bacteria.

Research Progress on Bioactive Factors against Skin Aging.

The relentless pursuit of effective strategies against skin aging has led to significant interest in the role of bioactive factors, particularly secondary metabolites from natural sources. The purpose of this study is to meticulously explore and summarize the recent advancements in understanding and utilization of bioactive factors against skin aging, with a focus on their sources, mechanisms of action, and therapeutic potential. Skin, the largest organ of the body, directly interacts with the external environment, making it susceptible to aging influenced by factors such as UV radiation, pollution, and oxidative stress. Among various interventions, bioactive factors, including peptides, amino acids, and secondary metabolites, have shown promising anti-aging effects by modulating the biological pathways associated with skin integrity and youthfulness. This article provides a comprehensive overview of these bioactive compounds, emphasizing collagen peptides, antioxidants, and herbal extracts, and discusses their effectiveness in promoting collagen synthesis, enhancing skin barrier function, and mitigating the visible signs of aging. By presenting a synthesis of the current research, this study aims to highlight the therapeutic potential of these bioactive factors in developing innovative anti-aging skin care solutions, thereby contributing to the broader field of dermatological research and offering new perspectives for future studies. Our findings underscore the importance of the continued exploration of bioactive compounds for their potential to revolutionize anti-aging skin care and improve skin health and aesthetics.

Research Progress in Skin Aging and Immunity.

Skin aging is a complex process involving structural and functional changes and is characterized by a decrease in collagen content, reduced skin thickness, dryness, and the formation of wrinkles. This process is underpinned by multiple mechanisms including the free radical theory, inflammation theory, photoaging theory, and metabolic theory. The skin immune system, an indispensable part of the body's defense mechanism, comprises macrophages, lymphocytes, dendritic cells, and mast cells. These cells play a pivotal role in maintaining skin homeostasis and responding to injury or infection. As age advances, along with various internal and external environmental stimuli, skin immune cells may undergo senescence or accelerated aging, characterized by reduced cell division capability, increased mortality, changes in gene expression patterns and signaling pathways, and altered immune cell functions. These changes collectively impact the overall function of the immune system. This review summarizes the relationship between skin aging and immunity and explores the characteristics of skin aging, the composition and function of the skin immune system, the aging of immune cells, and the effects of these cells on immune function and skin aging. Immune dysfunction plays a significant role in skin aging, suggesting that immunoregulation may become one of the important strategies for the prevention and treatment of skin aging.

Bif­1 inhibits activation of inflammasome through autophagy regulatory mechanism.

Inflammasome activation is a crucial mechanism in inflammatory responses. Bax­interacting factor 1 (Bif­1) is required for the normal formation of autophagosomes, but its ability to exert an inflammatory regulatory effect remains unclear. The aim of the present study was to explore the role of Bif­1 in inflammation, possibly mediated through autophagy regulation. Using a lipopolysaccharide (LPS)/adenosine triphosphate (ATP)­induced inflammatory model in J774A.1 cells, the effect of Bif­1 on inflammasome activation and the underlying mechanisms involving autophagy regulation were investigated. Elevated levels of NLR family pyrin domain containing protein 3 inflammasome and interleukin­1ß (IL­1ß) proteins were observed in J774A.1 cells after LPS/ATP induction. Furthermore, Bif­1 and autophagy activity were significantly upregulated in inflammatory cells. Inhibition of autophagy resulted in inflammasome activation. Silencing Bif­1 expression significantly upregulated IL­1ß levels and inhibited autophagy activity, suggesting a potential anti­inflammatory role of Bif­1 mediated by autophagy. Additionally, inhibition of the nuclear factor­κB (NF­κB) signaling pathway downregulated Bif­1 and inhibited autophagy activity, highlighting the importance of NF­κB in the regulation of Bif­1 and autophagy. In summary, the current study revealed that Bif­1 is a critical anti­inflammatory factor against inflammasome activation mediated by a mechanism of autophagy regulation, indicating its potential as a therapeutic target for inflammatory regulation.

Advancements in the Cultivation, Active Components, and Pharmacological Activities of Taxus mairei.

Taxus mairei (Lemée and H.Lév.) S.Y.Hu, indigenous to the southern regions of China, is an evergreen tree belonging to the genus Taxus of the Taxaceae family. Owing to its content of various bioactive compounds, it exhibits multiple pharmacological activities and has been widely applied in clinical medicine. This article comprehensively discusses the current state of cultivation, chemical constituents, applications in the pharmaceutical field, and the challenges faced by T. mairei. The paper begins by detailing the ecological distribution of T. mairei, aiming to provide an in-depth understanding of its origin and cultivation overview. In terms of chemical composition, the article thoroughly summarizes the extracts and monomeric components of T. mairei, unveiling their pharmacological activities and elucidating the mechanisms of action based on the latest scientific research, as well as their potential as lead compounds in new drug development. The article also addresses the challenges in the T. mairei research, such as the difficulties in extracting and synthesizing active components and the need for sustainable utilization strategies. In summary, T. mairei is a rare species important for biodiversity conservation and demonstrates significant research and application potential in drug development and disease treatment.

Toward grouped-reservoir computing: organic neuromorphic vertical transistor with distributed reservoir states for efficient recognition and prediction.

Reservoir computing has attracted considerable attention due to its low training cost. However, existing neuromorphic hardware, focusing mainly on shallow-reservoir computing, faces challenges in providing adequate spatial and temporal scales characteristic for effective computing. Here, we report an ultra-short channel organic neuromorphic vertical transistor with distributed reservoir states. The carrier dynamics used to map signals are enriched by coupled multivariate physics mechanisms, while the vertical architecture employed greatly increases the feedback intensity of the device. Consequently, the device as a reservoir, effectively mapping sequential signals into distributed reservoir state space with 1152 reservoir states, and the range ratio of temporal and spatial characteristics can simultaneously reach 2640 and 650, respectively. The grouped-reservoir computing based on the device can simultaneously adapt to different spatiotemporal task, achieving recognition accuracy over 94% and prediction correlation over 95%. This work proposes a new strategy for developing high-performance reservoir computing networks.

Temporal topic model for clinical pathway mining from electronic medical records.

BACKGROUND: In recent years, the discovery of clinical pathways (CPs) from electronic medical records (EMRs) data has received increasing attention because it can directly support clinical doctors with explicit treatment knowledge, which is one of the key challenges in the development of intelligent healthcare services. However, the existing work has focused on topic probabilistic models, which usually produce treatment patterns with similar treatment activities, and such discovered treatment patterns do not take into account the temporal process of patient treatment which does not meet the needs of practical medical applications. METHODS: Based on the assumption that CPs can be derived from the data of EMRs which usually record the treatment process of patients, this paper proposes a new CPs mining method from EMRs, an extended form of the traditional topic model - the temporal topic model (TTM). The method can capture the treatment topics and the corresponding treatment timestamps for each treatment day. RESULTS: Experimental research conducted on a real-world dataset of patients' hospitalization processes, and the achieved results demonstrate the applicability and usefulness of the proposed methodology for CPs mining. Compared to existing benchmarks, our model shows significant improvement and robustness. CONCLUSION: Our TTM provides a more competitive way to mine potential CPs considering the temporal features of the EMR data, providing a very prospective tool to support clinical diagnostic decisions.

An improved binary particle swarm optimization algorithm for clinical cancer biomarker identification in microarray data.

BACKGROUND AND OBJECTIVE: The limited number of samples and high-dimensional features in microarray data make selecting a small number of features for disease diagnosis a challenging problem. Traditional feature selection methods based on evolutionary algorithms are difficult to search for the optimal set of features in a limited time when dealing with the high-dimensional feature selection problem. New solutions are proposed to solve the above problems. METHODS: In this paper, we propose a hybrid feature selection method (C-IFBPFE) for biomarker identification in microarray data, which combines clustering and improved binary particle swarm optimization while incorporating an embedded feature elimination strategy. Firstly, an adaptive redundant feature judgment method based on correlation clustering is proposed for feature screening to reduce the search space in the subsequent stage. Secondly, we propose an improved flipping probability-based binary particle swarm optimization (IFBPSO), better applicable to the binary particle swarm optimization problem. Finally, we also design a new feature elimination (FE) strategy embedded in the binary particle swarm optimization algorithm. This strategy gradually removes poorer features during iterations to reduce the number of features and improve accuracy. RESULTS: We compared C-IFBPFE with other published hybrid feature selection methods on eight public datasets and analyzed the impact of each improvement. The proposed method outperforms other current state-of-the-art feature selection methods in terms of accuracy, number of features, sensitivity, and specificity. The ablation study of this method validates the efficacy of each component, especially the proposed feature elimination strategy significantly improves the performance of the algorithm. CONCLUSIONS: The hybrid feature selection method proposed in this paper helps address the issue of high-dimensional microarray data with few samples. It can select a small subset of features and achieve high classification accuracy on microarray datasets. Additionally, independent validation of the selected features shows that those chosen by C-IFBPFE have strong correlations with disease phenotypes and can identify important biomarkers from data related to biomedical problems.

Association between systemic lupus erythematosus and inflammatory bowel disease in European and East Asian populations: a two-sample Mendelian randomization study.

Background: Previous studies have shown a coexistence phenomenon between systemic lupus erythematosus (SLE) and inflammatory bowel disease (IBD), but the causal relationship between them is still unclear. Therefore, we conducted a two-sample Mendelian randomization (MR) analysis using publicly available summary statistics data to evaluate whether there was a causal relationship between the two diseases. Methods: Summary statistics for SLE and IBD were downloaded from the Open Genome-Wide Association Study and the International Inflammatory Bowel Disease Genetics Consortium. European and East Asian populations were included in this MR work. We adopted a series of methods to select instrumental variables that are closely related to SLE and IBD. To make the conclusion more reliable, we applied a variety of different analysis methods, among which the inverse variance-weighted (IVW) method was the main method. In addition, heterogeneity, pleiotropy, and sensitivity were assessed to make the conclusions more convincing. Results: In the European population, a negative causal relationship was observed between SLE and overall IBD (OR = 0.94; 95% CI = 0.90, 0.98; P < 0.004) and ulcerative colitis (UC) (OR = 0.93; 95% CI = 0.88, 0.98; P = 0.006). After removing outliers with Mendelian Randomization Pleiotropy RESidual Sum and Outlier (MR-PRESSO), the results remained consistent with IVW. However, there was no causal relationship between SLE and Crohn's disease. In the East Asian population, no causal relationship was found between SLE and IBD. Conclusion: Our results found that genetic susceptibility to SLE was associated with lower overall IBD risk and UC risk in European populations. In contrast, no association between SLE and IBD was found in East Asian populations. This work might enrich the previous research results, and it may provide some references for research in the future.

Research Progress in Skin Aging, Metabolism, and Related Products.

In recent years, skin aging has received increasing attention. Many factors affect skin aging, and research has shown that metabolism plays a vital role in skin aging, but there needs to be a more systematic review. This article reviews the interaction between skin metabolism and aging from the perspectives of glucose, protein, and lipid metabolism and explores relevant strategies for skin metabolism regulation. We found that skin aging affects the metabolism of three major substances, which are glucose, protein, and lipids, and the metabolism of the three major substances in the skin also affects the process of skin aging. Some drugs or compounds can regulate the metabolic disorders mentioned above to exert anti-aging effects. Currently, there are a variety of products, but most of them focus on improving skin collagen levels. Skin aging is closely related to metabolism, and they interact with each other. Regulating specific metabolic disorders in the skin is an important anti-aging strategy. Research and development have focused on improving collagen levels, while the regulation of other skin glycosylation and lipid disorders including key membrane or cytoskeleton proteins is relatively rare. Further research and development are expected.

Coal Structure Characteristics of the 2# Coal Seam in the Jiaozuo Mining Area and Its Geological Dependence.

To clarify the coal structure, spatial distribution, and controlling factors of the 2# coal seam in Jiaozuo mining, the drilling coal samples were collected to observe the coal type and coal structure. The coal macerals were identified by a MPVSP microscope photometer, and the spatial characteristics of the coal structure were obtained through interpreting deep lateral resistivity logging, natural gamma ray logging, density logging, and acoustic logging curves. The influence of coal properties, burial depth, geological stress, and faults on the coal structure were discussed correspondingly. The results exhibit that granulitic-mylonite coal was most developed in the 2# coal seam, followed by primary coal and cataclastic coal; the coal type was dominated by semibright coal, followed by clarain and semidull coal. Granulitic-mylonite, cataclastic, and primary coals were the main components of clarain, semibright coal, and semidull coal, respectively. Higher vitrinite and organic matter contents were conducive to the development of granulitic-mylonite. The coal structure combinations were spatially varied, and the granulitic-mylonite combinations were the most common. Granulitic-mylonite coal was developed in the east and south parts of the study area, and the coal structure was fragmented with a greater burial depth and larger thickness. The geological stress is the fundamental cause of coal structure damage as well as the cutting of faults.

Transformer-Based Multi-Modal Data Fusion Method for COPD Classification and Physiological and Biochemical Indicators Identification.

As the number of modalities in biomedical data continues to increase, the significance of multi-modal data becomes evident in capturing complex relationships between biological processes, thereby complementing disease classification. However, the current multi-modal fusion methods for biomedical data require more effective exploitation of intra- and inter-modal interactions, and the application of powerful fusion methods to biomedical data is relatively rare. In this paper, we propose a novel multi-modal data fusion method that addresses these limitations. Our proposed method utilizes a graph neural network and a 3D convolutional network to identify intra-modal relationships. By doing so, we can extract meaningful features from each modality, preserving crucial information. To fuse information from different modalities, we employ the Low-rank Multi-modal Fusion method, which effectively integrates multiple modalities while reducing noise and redundancy. Additionally, our method incorporates the Cross-modal Transformer to automatically learn relationships between different modalities, facilitating enhanced information exchange and representation. We validate the effectiveness of our proposed method using lung CT imaging data and physiological and biochemical data obtained from patients diagnosed with Chronic Obstructive Pulmonary Disease (COPD). Our method demonstrates superior performance compared to various fusion methods and their variants in terms of disease classification accuracy.

Cytoplasm Hydrogelation-Mediated Cardiomyocyte Sponge Alleviated Coxsackievirus B3 Infection.

Viral myocarditis (VMC), commonly caused by coxsackievirus B3 (CVB3) infection, lacks specific treatments and leads to serious heart conditions. Current treatments, such as IFNα and ribavirin, show limited effectiveness. Herein, rather than inhibiting virus replication, this study introduces a novel cardiomyocyte sponge, intracellular gelated cardiomyocytes (GCs), to trap and neutralize CVB3 via a receptor-ligand interaction, such as CAR and CD55. By maintaining cellular morphology, GCs serve as sponges for CVB3, inhibiting infection. In vitro results revealed that GCs could inhibit CVB3 infection on HeLa cells. In vivo, GCs exhibited a strong immune escape ability and effectively inhibited CVB3-induced viral myocarditis with a high safety profile. The most significant implication of this study is to develop a universal antivirus infection strategy via intracellular gelation of the host cell, which can be employed not only for treating defined pathogenic viruses but also for a rapid response to infection outbreaks caused by mutable and unknown viruses.

Antibacterial Activity and Mechanism of Canagliflozin against Methicillin-Resistant Staphylococcus aureus.

Diabetic foot infection (DFI) is a common complication in diabetes patients, with foot infections being the leading cause of amputations. Staphylococcus aureus is frequently found in diabetic foot infections, of which methicillin-resistant Staphylococcus aureus (MRSA) has become a major clinical and epidemiological challenge. Since MRSA strains are resistant to most ß-lactam antibiotics, and also partially resistant to other antibiotics, treatment is difficult and costly. The emergence of drug-resistant bacteria often arises from overuse or misuse of antibiotics. Clinically, canagliflozin is commonly used for the treatment of type 2 diabetes. On this basis, we investigated the antibacterial activity and mechanism of canagliflozin against MRSA, with the aim to discover novel functions of canagliflozin and provide new insights for the treatment of MRSA. Using the microbroth dilution method to determine the half maximal inhibitory concentration of drugs, we found that canagliflozin not only can inhibit the growth of methicillin-sensitive Staphylococcus aureus (MSSA) but also exhibits antibacterial activity against MRSA. The IC50 values, at approximately 56.01 µM and 57.60 µM, were almost the same. At 12 h, canagliflozin showed a significant antibacterial effect against MRSA at and above 30 µM. In addition, its combined use with penicillin achieved better antibacterial effects, which were increased by about three times. Additive antibacterial activity (FICI = 0.69) was found between penicillin and canagliflozin, which was better than that of doxycycline and canagliflozin (FICI = 0.95). Canagliflozin also affected bacterial metabolic markers, such as glucose, ATP, and lactic acid. The results of crystal violet staining indicate that canagliflozin disrupted the formation of bacterial biofilm. Our electron microscopy results showed that canagliflozin distorted the bacterial cell wall. The results of RT-PCR suggest that canagliflozin down-regulated the expressions of biofilm-related gene (clfA, cna, agrC, mgrA, hld) and methicillin-resistance gene (mecA), which was related to MRSA. Molecular docking also indicated that canagliflozin affected some interesting targets of MRSA, such as the sarA, crtM and fnbA proteins. In conclusion, canagliflozin exhibits antibacterial activity against MRSA by affecting bacterial metabolism, inhibiting its biofilm formation, distorting the bacterial cell wall, and altering the gene expression of biofilm formation and its virulence. Our study reveals the antibacterial activity of canagliflozin against MRSA, providing a new reference for treating diabetic foot infections.

A quantitative heterogeneity analysis approach to molecular subtype recognition of breast cancer in dynamic contrast-enhanced magnetic imaging images from radiomics data.

Background: Breast cancer is a major cause of mortality among women worldwide. Dynamic contrast-enhanced breast magnetic resonance imaging (DCE-MRI) is a good imaging technique that can show temporal information about the kinetics of the contrast agent in suspicious breast lesions as well as acceptable spatial resolution. Computer-aided detection systems assist in the detection of lesions through medical image processing techniques combined with computerized analysis and calculation, which in turn helps radiologists recognize molecular subtypes of breast lesions that will be beneficial for better treatment plan decisions. Methods: In this paper, a computer-aided diagnosis method is proposed to automatically locate breast cancer lesions and identify molecular subtypes of breast cancer with heterogeneity analysis from radiomics data. A fast region-based convolutional network (Faster R-CNN) framework is first applied to images to detect breast cancer lesions. Then, the heterogeneous regions of every breast cancer lesion are extracted. Based on the multiple visual and kinetic radiomics features extracted from the heterogeneous regions, a temporal bag of visual word model is proposed, which takes into account the dynamic characteristics of both lesion and heterogeneous regions in images over time. The recognition task of molecular subtypes of breast lesions is realized based on a stacking classification model. Results: At the genetic level, breast cancer is divided into four molecular subtypes, namely, luminal epithelial type A (Luminal A), luminal epithelial type B (Luminal B), HER-2 overexpression and basal cell type. The experimental results show that the precision of the four subtypes is 93%, 94%, 83%, 86%; the recall is 96%, 80%, 91%, 94%; and the F1-score is 95%, 86%, 87%. Conclusions: The experimental results denote the influence of heterogeneous regions on the recognition task. The DCE-MRI-based approach to identify molecular typing of breast cancer for noninvasive diagnosis will contribute to the development of breast cancer treatment, improved outcomes and reduced mortality.

Research Progress on Skin Aging and Active Ingredients.

With the advancement of living standards in modern society and the emergence of an aging population, an increasing number of people are becoming interested in the topic of aging and anti-aging. An important feature of aging is skin aging, and women are particularly concerned about skin aging. In the field of cosmetics, the market share of anti-aging products is increasing year by year. This article reviews the research and development progress of skin aging and related active compounds both domestically and internationally in recent years. The results show that, in terms of the research on skin aging, the popular theories mainly include free radicals and oxidative stress theory, inflammation theory, photoaging theory, and nonenzymatic glycosyl chemistry theory. In terms of research on the active ingredients with anti-aging activities in the skin, there are numerous reports on related products in clinical studies on human subjects, animal experiments, and experimental studies on cell cultures, with a variety of types. Most of the compounds against skin aging are sourced from natural products and their action mechanisms are mainly related to scavenging oxygen free radicals and enhancing antioxidant defenses. This review provides important references for the future research of skin aging and the development of related products. Although there is a great progress in skin aging including related active ingredients, ideal compounds or products are still lacking and need to be further validated. New mechanisms of skin aging, new active ingredients sourced from natural and artificial products, and new pharmaceutical forms including further clinical validations should be further investigated in the future.

Dual-GCN-based deep clustering with triplet contrast for ScRNA-seq data analysis.

Single-cell RNA sequencing (ScRNA-seq) technology reveals gene expression information at the cellular level. The critical tasks in ScRNA-seq data analysis are clustering and dimensionality reduction. Recent deep clustering algorithms are used to optimize the two tasks jointly, and their variations, graph-based deep clustering algorithms, are used to capture and preserve topological information in the process. However, the existing graph-based deep clustering algorithms ignore the distribution information of nodes when constructing cell graphs which leads to incomplete information in the embedding representation; and graph convolutional networks (GCN), which are most commonly used, often suffer from over-smoothing that leads to high sample similarity in the embedding representation and then poor clustering performance. Here, the dual-GCN-based deep clustering with Triplet contrast (scDGDC) is proposed for dimensionality reduction and clustering of scRNA-seq data. Two critical components are dual-GCN-based encoder for capturing more comprehensive topological information and triplet contrast for reducing GCN over-smoothing. The two components improve the dimensionality reduction and clustering performance of scDGDC in terms of information acquisition and model optimization, respectively. The experiments on eight real ScRNA-seq datasets showed that scDGDC achieves excellent performance for both clustering and dimensionality reduction tasks and is high robustness to parameters.