Identification of key genes related to heart failure by analysis of expression profiles.

Introduction: To elucidate the candidate biomarkers involved in the pathogenesis process of heart failure (HF) via analysis of differentially expressed genes (DEGs) of the dataset from the Gene Expression Omnibus (GEO). Material and methods: The GSE76701 gene expression profiles regarding the HF and control subjects were respectively analysed. Briefly, DEGs were firstly identified and subjected to Cytoscape plug-in ClueGO + CluePedia and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses. A protein-protein interaction (PPI) network was then built to analyse the interaction between DEGs, followed by the construction of an interaction network by combining with hub genes with the targeted miRNA genes of DEGs to identify the key molecules of HF. In addition, potential drugs targeting key DEGs were sought using the drug-gene interaction database (DGIdb), and a drug-mRNA-miRNA interaction network was also constructed. Results: A total of 489 DEGs were verified between HF and control, which mainly enriched in type I interferon and leukocyte migration according to molecular function. Significantly increased levels of GAPDH, GALM1, MMP9, CCL5, and GNAL2 were found in the HF setting and were identified as the hub genes based on the PPI network. Furthermore, according to the drug-mRNA-miRNA network, FCGR2B, CCND1, and NF-κb, as well as corresponding miRNA-605-5p, miRNA-147a, and miRNA-671-5p were identified as the drug targets of HF. Conclusions: The hub genes GAPDH, GALM1, MMP9, CCL5, and GNAL2 were significantly increased in HF. miRNA-605-5p, miRNA-147a, and miRNA-671-5p were predicted as the drug target-interacted gene-miRNA of HF.

Insulin resistance and coronary inflammation in patients with coronary artery disease: a cross-sectional study.

BACKGROUND: Insulin resistance (IR) is associated with coronary artery disease (CAD) severity. However, its underlying mechanisms are not fully understood. Therefore, our study aimed to explore the relationship between IR and coronary inflammation and investigate the synergistic and mediating effects of coronary inflammation on the association between IR and CAD severity. METHODS: Consecutive patients with CAD who underwent coronary angiography and coronary computed tomography angiography between April 2018 and March 2023 were enrolled. The triglyceride-glucose index (TyG index) and peri-coronary adipose tissue (PCAT) attenuation around the proximal right coronary artery (RCA) were used to evaluate IR and coronary inflammation, respectively. The correlation between the TyG index and PCAT attenuation was analyzed using linear regression models. Logistic regression models were further used for investigating the correlation of the TyG index and PCAT attenuation with CAD severity. A mediation analysis assessed the correlation between IR and CAD severity mediated by coronary inflammation. RESULTS: A total of 569 participants (mean age, 62 ± 11 years; 67.8% men) were included in the study. PCAT attenuation was positively associated with the TyG index (r = 0.166; P < 0.001). After adjusting for potential confounders, the per standard deviation increment in the TyG index was associated with a 1.791 Hounsfield unit (HU) increase (95% confidence interval [CI], 0.920-2.662 HU; P < 0.001) in the PCAT attenuation. In total, 382 (67.1%) patients had multivessel CAD. The patients in the high-TyG index/high PCAT attenuation group had approximately 3.2 times the odds of multivessel CAD compared with those in the low-TyG index/low PCAT attenuation group (odds ratio, 3.199; 95%CI, 1.826-5.607; P < 0.001). Mediation analysis indicated that PCAT attenuation mediated 31.66% of the correlation between the TyG index and multivessel CAD. CONCLUSIONS: The TyG index positively correlated with PCAT attenuation in patients with CAD. The TyG index and PCAT attenuation showed a synergistic correlation with multivessel CAD. Furthermore, PCAT attenuation partially mediated the relationship between the TyG index and CAD severity. Controlling inflammation in patients with high IR and coronary inflammation may provide additional benefits.

Caspase-3/gasdermin-E axis facilitates the progression of coronary artery calcification by inducing the release of high mobility group box protein 1.

Coronary artery calcification (CAC) is commonly observed in atherosclerotic plaques, which is a pathogenic factor for severe coronary artery disease (CAD). The phenotype changes of vascular smooth muscle cells (VSMCs) are found to participate in CAC progression, which is mainly induced by vascular inflammation and oxidative stress (OS). HMGB1, a critical inflammatory cytokine, is recently reported to induce arterial calcification, which is regulated by the Caspase-3/gasdermin-E (GSDME) axis. However, the function of the Caspase-3/GSDME axis in CAC is unknown. Herein, the involvement of the Caspase-3/GSDME axis in CAC was studied to explore the possible targets for CAC. CAC model was constructed in mice, which was verified by red cytoplasm in coronary artery tissues, increased macrophage infiltration, aggravated inflammation, and enhanced RAGE signaling, accompanied by an increased release of HMGB1 and an activated Caspase-3/ GSDME axis. In ß-GP-treated MOVAS-1 cells, calcification, the ROS accumulation, enhanced LDH and HMGB1 release, enlarged macrophage production, aggravated inflammation, and activated RAGE signaling were observed, which were markedly abolished by the transfection of si-HMGB1 and si-GSDME. Moreover, the calcification deposition, the activity of Caspase-3/ GSDME axis, release of HMGB1, macrophage infiltration, cytokine production, and RAGE signaling in CAC mice were notably alleviated by VSMCs-specific GSDME knockdown, not by hematopoietic stem cells (HSCs)-specific GSDME knockdown. Collectively, Caspase-3/GSDME axis facilitated the progression of CAC by inducing the release of HMGB1.

Feature Fusion Based on Graph Convolution Network for Modulation Classification in Underwater Communication.

Automatic modulation classification (AMC) of underwater acoustic communication signals is of great significance in national defense and marine military. Accurate modulation classification methods can make great contributions to accurately grasping the parameters and characteristics of enemy communication systems. While a poor underwater acoustic channel makes it difficult to classify the modulation types correctly. Feature extraction and deep learning methods have proven to be effective methods for the modulation classification of underwater acoustic communication signals, but their performance is still limited by the complex underwater communication environment. Graph convolution networks (GCN) can learn the graph structured information of the data, making it an effective method for processing structured data. To improve the stability and robustness of AMC in underwater channels, we combined the feature extraction and deep learning methods by fusing the multi-domain features and deep features using GCN. The proposed method takes the relationships among the different multi-domain features and deep features into account. Firstly, a feature graph was built using the properties of the features. Secondly, multi-domain features were extracted from the received signals and deep features were extracted from the signals using a deep neural network. Thirdly, we constructed the input of GCN using these features and the graph. Then, the multi-domain features and deep features were fused by the GCN. Finally, we classified the modulation types using the output of GCN by way of a softmax layer. We conducted the experiments on a simulated dataset and a real-world dataset, respectively. The results show that the AMC based on GCN can achieve a significant improvement in performance compared to the current state-of-the-art methods. Our approach is robust in underwater acoustic channels.

Molecular structure-dependent contribution of reactive species to organic pollutant degradation using nanosheet Bi2Fe4O9 activated peroxymonosulfate.

Iron-based catalysts have attracted increasing attention in heterogeneous activation of peroxymonosulfate (PMS). However, the activity of most iron-based heterogenous catalysts is not satisfactory for practical application and the proposed activation mechanisms of PMS by iron-based heterogenous catalyst vary case by case. This study prepared Bi2Fe4O9 (BFO) nanosheet with super high activity toward PMS, which was comparable to its homogeneous counterpart at pH 3.0 and superior to its homogeneous counterpart at pH 7.0. Fe sites, lattice oxygen and oxygen vacancies on BFO surface were believed to be involved in the activation of PMS. By using electron paramagnetic resonance (EPR), radical scavenging tests, 57Fe Mössbauer and 18O isotope-labeling technique, the generation of reactive species including sulfate radicals, hydroxyl radicals, superoxide and Fe (IV) were confirmed in BFO/PMS system. However, the contribution of reactive species to the elimination of organic pollutants very much depends on their molecular structure. The effect of water matrices on the elimination of organic pollutants also hinges on their molecular structure. This study implies that the molecular structure of organic pollutants governs their oxidation mechanism and their fate in iron-based heterogeneous Fenton-like system and further broadens our knowledge on the activation mechanism of PMS by iron-based heterogeneous catalyst.

The application of metagenomic next-generation sequencing in patients with infection or colonization caused by Lichtheimia species.

Background: Mucormycosis is considered the fourth most common invasive fungal disease after candidiasis, aspergillosis and cryptococcosis. Lichtheimia species accounted for 5%-29% of all mucormycosis. However, available data on species-specific analysis of Lichtheimia infections are limited. Methods: This study included nine patients hospitalized in five hospitals in two cities in south China with mucormycosis or colonization caused by Lichtheimia species, diagnosed mainly by metagenomic next-generation sequencing (mNGS). The corresponding medical records were reviewed, and the clinical data analyzed included demographic characteristics, site of infection, host factors and type of underlying disease, diagnosis, clinical course, management, and prognosis. Results: In this study, nine patients with Lichtheimia infections or colonization had a recent history of haematological malignancy (33.3%), solid organ transplants (33.3%), pulmonary disease (22.2%), and trauma (11.1%) and were categorized as 11.1% (one case) proven, 66.7% (six cases) probable mucormycosis and 22.2% (two cases) colonization. Pulmonary mucormycosis or colonization was the predominant presentation in 77.8% of cases and mucormycosis caused by Lichtheimia resulted in death in four out of seven patients (57.1%). Conclusion: These cases highlight the importance of early diagnosis and combined therapy for these sporadic yet life-threatening infections. Further studies on the diagnosis and control of Lichtheimia infection in China are required.

Automatic Modulation Classification for Underwater Acoustic Communication Signals Based on Deep Complex Networks.

Automatic modulation classification (AMC) is an important method for monitoring and identifying any underwater communication interference. Since the underwater acoustic communication scenario is full of multi-path fading and ocean ambient noise (OAN), coupled with the application of modern communication technology, which is usually susceptible to environmental influences, automatic modulation classification (AMC) becomes particularly difficult when it comes to an underwater scenario. Motivated by the deep complex networks (DCN), which have an innate ability to process complex data, we explore DCN for AMC of underwater acoustic communication signals. To integrate the signal processing method with deep learning and overcome the influences of underwater acoustic channels, we propose two complex physical signal processing layers based on DCN. The proposed layers include a deep complex matched filter (DCMF) and deep complex channel equalizer (DCCE), which are designed to remove noise and reduce the influence of multi-path fading for the received signals, respectively. Hierarchical DCN is constructed using the proposed method to achieve better performance of AMC. The influence of the real-world underwater acoustic communication scenario is taken into account; two underwater acoustic multi-path fading channels are conducted using the real-world ocean observation dataset, white Gaussian noise, and real-world OAN are used as the additive noise, respectively. Contrastive experiments show that the AMC based on DCN can achieve better performance than the traditional deep neural network based on real value (the average accuracy of the DCN is 5.3% higher than real-valued DNN). The proposed method based on DCN can effectively reduce the influence of underwater acoustic channels and improve the AMC performance in different underwater acoustic channels. The performance of the proposed method was verified on the real-world dataset. In the underwater acoustic channels, the proposed method outperforms a series of advanced AMC method.

Fe Single-Atom Catalyst for Cost-Effective yet Highly Efficient Heterogeneous Fenton Catalysis.

High energy consumption in pyrolyzing precursors for catalyst preparation would limit the application of nitrogen-doped carbon-based single-atom catalysts in actual pollutant remediation. Herein, we report an Fe single atom (7.67 wt %) loaded polyaniline catalyst (Fe-PANI) prepared via a simple impregnation process without pyrolysis. Both experimental characterizations and density functional theory calculations demonstrated that isolated -N═ group sites can fasten Fe atoms through Fe-N coordination in PANI, leading to a high stability of Fe atoms in a heterogeneous Fenton reaction. Highly dispersive yet dense -N═ groups in PANI can be protonated to be adsorption sites, which largely reduce the migration distance between reactive radicals and organics. More significantly, frontier molecular orbitals and spin-density distributions reveal that electrons can transfer from reduction groups of PANI to an Fe(III) site to accelerate its reduction. As a result, a remarkably boosted degradation behavior of organics under near-neutral conditions (pH 6), with low H2O2 concentration, was achieved. This cost-effective Fe-PANI catalyst with high catalytic activity, stability, and adsorption performance has great potential for industrial-level wastewater treatment.

Tailoring co-assembly loading of doxorubicin in solvent-triggering gel.

Noncovalent interactions are ubiquitous, endowing high feasibility on assembly and disassembly of gel network structure. Loading anticancer drugs in low molecular weight gelator (LMWG)-based gel through a noncovalently co-assembly process shows advantages of high efficacy, thixotropy, and controllable release. Drug-loaded fluorenylmethyloxycarbonyl-phenylalanine (Fmoc-F)/DMSO/H2O-doxorubicin (DOX) gels were fabricated by an effective solvent-triggering method dominated by solvated Fmoc-F with DMSO. Density Functional Theory (DFT) calculation results show that the noncovalent interactions between Fmoc-F and DOX drive the co-assembly of the gel. DOX can assemble with Fmoc-F and realize its co-assembly loading through the H-bonding and π-π stacking, similar to the way that gel networks form. Depending on a network dis-assembly process, sustained release of DOX was achieved along with carrier decomposition through a repetitive diffusion-surface erosion process. DOX loading and release prove the non-covalent interactions and the mechanism for controlling the assembly process. By such tailoring co-assembled loading, the administration of DOX is hoped to be optimized to improve the clinical application.

Domination of H-Bond Interactions in the Solvent-Triggering Gelation Process.

Gels prepared with the solvent-triggering method are attractive for their easy and fast preparation; however, the role of solvents in this process remains unclear, which hinders the efficient and accurate control of desired gel properties. In this study, the role of solvents in the solvent-triggering gelation process is studied using 9-fluorenylmethoxycarbonyl (Fmoc)-protected diphenylalanine (Fmoc-FF) as the gelator. Density functional theory (DFT)-based calculations and corresponding wavefunction analyses are conducted to identify the H-bonding interaction sites between the molecules. The calculation results clearly annotate the activating role of DMF and the triggering role of H2O in the gelation process. The solvation of Fmoc-FF by DMF can activate the H-bonding sites on the peptide chain, showing a conformation reversal and higher electrostatic potentials. Then, the H-bonding between Fmoc-FF and H2O is facilitated to trigger gelation. The physical Fmoc-FF/DMF/H2O gels show easily tuned mechanical strengths (G' of 102-105 Pa), injectable potentials (general yield strain < 100%), and stable recoverability (80-98% within 100 s). The regulation of these properties depends on not only the gelator concentration but also the H-bonding interactions with solvent molecules, which have seldom been studied in detail before. By understanding the effect of solvents, low-molecular-weight gelator-based gels can be designed, prepared, and tuned efficiently for potential applications.

Histone deacetylase HDAC4 participates in the pathological process of myocardial ischemia-reperfusion injury via MEKK1/JNK pathway by binding to miR-206.

Histone deacetylases (HDACs) and microRNAs (miRs) have been reported to exert pivotal roles on the pathogenesis of myocardial ischemia-reperfusion injury (MIRI). Therefore, the present study was performed to define the underlying role of HDAC4 and miR-206 in the pathological process of MIRI. An IRI rat model was established. The interaction between HDAC4 and the promoter region of miR-206 was determined using ChIP, and that between miR-206 and mitogen-activated protein kinase kinase kinase 1 (MEKK1) was determined using dual luciferase reporter gene assay. After the loss- or gain-of-function assay in cardiomyocytes, western blot analysis, RT-qPCR, TUNEL, and ELISA assay were performed to define the roles of HDAC4, miR-206, and MEKK1. Up-regulation of HDAC4 and down-regulation of miR-206 occurred in rat myocardial tissues and cardiomyocytes in MIRI. HDAC4 down-regulation or miR-206 up-regulation contributed to reduced cell apoptosis and the levels of tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), and malondialdehyde (MDA), while elevating the superoxide dismutase (SOD) and glutathione (GSH) contents. Meanwhile, HDAC4 silencing promoted the expression of miR-206, which targeted and negatively regulated MEKK1. Then inhibition of JNK phosphorylation reduced the cardiomyocyte apoptosis to alleviate MIRI. Coherently, HDAC4 silencing could up-regulate the expression of miR-206 to reduce cardiomyocyte apoptosis and inhibit oxidative stress, and exerting a protective effect on MIRI via the MEKK1/JNK pathway.

Green synthesis of poly(pyrrole methane) for enhanced adsorption of anionic and cationic dyes from aqueous solution.

The presence and accumulation of dyestuff in the environment is posing great harm to human beings. In this study, a novel poly(pyrrole methane) (PPm) adsorbent with abundant OH was greenly synthesized via a facile polymerization method. Its physicochemical properties were characterized in detail. Furthermore, the adsorption performance of PPm for anionic dye (acid red G, ARG) and cationic dye (methylene blue, MB) was comparatively studied with a typical dye adsorbent (polyprrrole, PPy). The results revealed that the adsorption of ARG or MB onto PPm followed pseudo-second-order model and Langmuir mode. The adsorption processes were endothermic and spontaneous. The maximum capacities of PPm to adsorb ARG and MB were 555.56 mg/g and 99.11 mg/g, which were about 10 and 2 times higher than that of PPy, respectively. PPm could be reused for 5 cycles without a significant decrease of its adsorption rate. The adsorption of ARG and MB is mainly attributed to electrostatic interaction and hydrogen bonding between ARG or MB and OH in PPm. Additionally, ARG could be adsorbed by ion exchange with the doped Cl- in PPm. Therefore, this study provides a new strategy to synthesis efficient adsorbent for the removal of both anionic and cationic dyes.

MMP9, CXCR1, TLR6, and MPO participant in the progression of coronary artery disease.

Coronary artery disease (CAD) is the most frequent cardiovascular disease, which is induced by the decreased myocardial blood supply. The present study is conducted to understand the mechanisms of CAD. The GSE98583, GSE69587, and GSE71226 datasets from the Gene Expression Omnibus database were obtained. The differentially expressed genes (DEGs) were analyzed by the limma package, then the DEGs appeared in two or three datasets were selected as the coregulated genes using the VENNY tool, followed by enrichment analysis using DAVID tool. Protein-protein interaction (PPI) network, microRNA-transcription factor-target regulatory network, and drug-gene network were visualized. Finally, quantitative PCR and dual-luciferase reporter assay were conducted to validate the expression of key genes and the target relationship. There were 221 coregulated genes in GSE98583, GSE69587, and GSE71226. Besides, four pathways and 23 functional terms for co-upregulated genes, and 11 functional terms for co-downregulated genes were enriched. The degrees of PPI network nodes matrix metallopeptidase 9 (MMP9), C-X-C motif chemokine receptor 1 (CXCR1), toll-like receptor 6 (TLR6), and myeloperoxidase (MPO) were relatively higher. Moreover, MPO could interact with MMP9, CXCR1, and TLR6 in the PPI network. In the regulatory network, TLR6 and MMP9 separately were targeted by miR-3960 and v-rel avian reticuloendotheliosis viral oncogene homolog A (RELA). Additionally, MMP9, CXCR1, and MPO were involved in the drug-gene network. The expression of MMP9, CXCR1, TLR6, and MPO were significantly upregulated in CAD samples than control, and miR-3960 could bind to TLR6 to inhibit its expression. CXCR1 and MPO might be involved in the progression of CAD. Besides, miR-3960 might function in the pathogenesis of CAD through targeting TLR6, and RELA might exert its role in CAD via targeting MMP9.

MicroRNA-2861 and microRNA-5115 regulates myocardial ischemia-reperfusion injury through the GPR30/mTOR signaling pathway by binding to GPR30.

Myocardial ischemia-reperfusion (I/R) injury, a major contributor to morbidity and mortality, represents a combination of intrinsic cellular response to ischemia and the extrinsic acute inflammatory response. In the present study, microarray analysis of GSE67308 and GSE50885 identified differentially expressed GPR30 and upstream regulatory miR-2861 and miR-5115 in myocardial I/R. Furthermore, GPR30 was confirmed as a common target gene of miR-2861 and miR-5115, and miR-2861 and miR-5115 inhibited GPR30 expression. Poor expression of GPR30 was identified in the myocardial I/R injury mouse model. Overexpressed GPR30 led to alleviated the pathological conditions, diminished myocardial infarct size and apoptosis of myocardial tissue in mice. Moreover, miR-2861 and miR-5115 were found to be highly expressed in the myocardial I/R injury mouse model and to subsequently accelerate the disease progression. Notably, PR30 curtailed the development of myocardial I/R injury through activation of the mTOR signaling pathway. The key findings suggested that miR-2861 and miR-5115 blocked the activation of the GPR30/mTOR signaling pathway by targeting GPR30, thereby accelerating myocardial I/R injury in mice.

Ship Type Classification by Convolutional Neural Networks with Auditory-like Mechanisms.

Ship type classification with radiated noise helps monitor the noise of shipping around the hydrophone deployment site. This paper introduces a convolutional neural network with several auditory-like mechanisms for ship type classification. The proposed model mainly includes a cochlea model and an auditory center model. In cochlea model, acoustic signal decomposition at basement membrane is implemented by time convolutional layer with auditory filters and dilated convolutions. The transformation of neural patterns at hair cells is modeled by a time frequency conversion layer to extract auditory features. In the auditory center model, auditory features are first selectively emphasized in a supervised manner. Then, spectro-temporal patterns are extracted by deep architecture with multistage auditory mechanisms. The whole model is optimized with an objective function of ship type classification to form the plasticity of the auditory system. The contributions compared with an auditory inspired convolutional neural network include the improvements in dilated convolutions, deep architecture and target layer. The proposed model can extract auditory features from a raw hydrophone signal and identify types of ships under different working conditions. The model achieved a classification accuracy of 87.2% on four ship types and ocean background noise.

Atherosclerosis-associated endothelial cell apoptosis by miRNA let7-b-mediated downregulation of HAS-2.

MicroRNAs (miRNAs) play essential roles in the regulation and pathophysiology of various types of human diseases including atherosclerosis. Increasing numbers of miRNAs have been identified to be important regulators in the progression of atherosclerosis by regulating gene expression. However, functional miRNAs and the underlying mechanisms involved in atherosclerosis need fully elucidation. In the present study, the function of miRNA let-7b was investigated in human aortic endothelial cells (HAECs). The results showed that downregulation of let-7b in the high-fat diet mice and HAECs was inversely correlated with the expression level of HAS-2. upregulation of let-7b significantly reduced apoptosis of HAECs. The results also revealed that HAS-2 was a target gene of let-7b and HAS-2 reduction reversed the antiapoptotic effect of let-7b through regulation of the P13K/Akt pathway. These results together suggest the potential of regulating the let-7b expression and endothelial apoptosis against development and progression of atherosclerosis.

A Deep Convolutional Neural Network Inspired by Auditory Perception for Underwater Acoustic Target Recognition.

Underwater acoustic target recognition (UATR) using ship-radiated noise faces big challenges due to the complex marine environment. In this paper, inspired by neural mechanisms of auditory perception, a new end-to-end deep neural network named auditory perception inspired Deep Convolutional Neural Network (ADCNN) is proposed for UATR. In the ADCNN model, inspired by the frequency component perception neural mechanism, a bank of multi-scale deep convolution filters are designed to decompose raw time domain signal into signals with different frequency components. Inspired by the plasticity neural mechanism, the parameters of the deep convolution filters are initialized randomly, and the is n learned and optimized for UATR. The n, max-pooling layers and fully connected layers extract features from each decomposed signal. Finally, in fusion layers, features from each decomposed signal are merged and deep feature representations are extracted to classify underwater acoustic targets. The ADCNN model simulates the deep acoustic information processing structure of the auditory system. Experimental results show that the proposed model can decompose, model and classify ship-radiated noise signals efficiently. It achieves a classification accuracy of 81.96%, which is the highest in the contrast experiments. The experimental results show that auditory perception inspired deep learning method has encouraging potential to improve the classification performance of UATR.

Fluorescent monitoring of the reaction kinetics of nonfluorescent molecules enabled by a fluorescent receptor.

A facile fluorescent method was developed to quantitatively monitor the hydrolysis kinetics of nonfluorescent esters by using a fluorecent endo-functionalized molecular tube and its recognition ability towards small polar molecules in water. It is possible to determine the apparent rate constants and study the structure-activity relationship.

Potential biomarkers for heart failure.

In this study, we identified candidate biomarkers for heart failure (HF). The gene expression profile GSE57338, containing 117 ischemic cardiomyopathic HF and 136 control samples, was downloaded and analyzed using various bioinformatics approaches. In total, 376 differentially expressed genes (DEGs) were identified, and four modules were explored in protein-protein interaction networks. DEGs (including ankyrin repeat and SOCS box-containing 14 [ASB14]) in the modules were mainly categorized by the function. Several relationships including interferon regulatory factor 1 (IRF1)-C-C motif chemokine ligand 5 (CCL5) were revealed in the transcription factor microRNA target gene regulatory network. Gene-drug analysis revealed 11 DEGs (such as the cluster of differentiation 163 [CD163]) for the target drugs. Data verification analysis identified 118 overlapping DEGs including ASB14, CD163, and CCL5. ASB14 may be involved in HF progression via protein ubiquitination and CCL5 may be involved in HF via the IRF1-CCL5 interaction. Genes including CD163 are potential biomarkers for HF.

Competitive Deep-Belief Networks for Underwater Acoustic Target Recognition.

Underwater acoustic target recognition based on ship-radiated noise belongs to the small-sample-size recognition problems. A competitive deep-belief network is proposed to learn features with more discriminative information from labeled and unlabeled samples. The proposed model consists of four stages: (1) A standard restricted Boltzmann machine is pretrained using a large number of unlabeled data to initialize its parameters; (2) the hidden units are grouped according to categories, which provides an initial clustering model for competitive learning; (3) competitive training and back-propagation algorithms are used to update the parameters to accomplish the task of clustering; (4) by applying layer-wise training and supervised fine-tuning, a deep neural network is built to obtain features. Experimental results show that the proposed method can achieve classification accuracy of 90.89%, which is 8.95% higher than the accuracy obtained by the compared methods. In addition, the highest accuracy of our method is obtained with fewer features than other methods.