Search

Abstract Non-coding RNAs (ncRNAs) draw much attention from studies widely in recent years because they play vital roles in life activities. As a good complement to wet experiment methods, computational prediction methods can greatly save experimental costs. However, high false-negative data and insufficient use of multi-source information can affect the performance of computational prediction methods. Furthermore, many computational methods do not have good robustness and generalization on different datasets. In this work, we propose an effective end-to-end computing framework, called GDCL-NcDA, of deep graph learning and deep matrix factorization (DMF) with contrastive learning, which identifies the latent ncRNA-disease association on diverse multi-source heterogeneous networks (MHNs). The diverse MHNs include different similarity networks and proven associations among ncRNAs (miRNAs, circRNAs, and lncRNAs), genes, and diseases. Firstly, GDCL-NcDA employs deep graph convolutional network and multiple attention mechanisms to adaptively integrate multi-source of MHNs and reconstruct the ncRNA-disease association graph. Then, GDCL-NcDA utilizes DMF to predict the latent disease-associated ncRNAs based on the reconstructed graphs to reduce the impact of the false-negatives from the original associations. Finally, GDCL-NcDA uses contrastive learning (CL) to generate a contrastive loss on the reconstructed graphs and the predicted graphs to improve the generalization and robustness of our GDCL-NcDA framework. The experimental results show that GDCL-NcDA outperforms highly related computational methods. Moreover, case studies demonstrate the effectiveness of GDCL-NcDA in identifying the associations among diversiform ncRNAs and diseases.

With the development of communication and monitoring techniques, multi-source information in form of multi-view data is expected to support fault-cause identification for transmission line maintenance and fault disposal. However, the combined used of multi-view data poses a great challenge of information fusion. In respond to this challenge, this paper proposes a multi-view sparse feature selection (MVSFS) method to combine contextual and waveform features for fault-cause classification. l2,1-norm sparsity regularization is adopted to select discriminative features across two views and an ɛ-dragging technique is integrated into the regression model to enhance its discriminant ability. Subsequently, an iteration optimization algorithm is devised to solve the model. Experiments on a real-life fault dataset demonstrate that contextual and waveform features can achieve higher accuracy than single view learning only through appropriate fusion, and the proposed MVSFS can overcome the drawback of conventional feature selection methods and make a great improvement of classification performance with multi-view data.

In the study of transmission line fault location, most of the previous artificial intelligence-based location methods rely heavily on feature extraction of fault signals, which depend on the researcher’s level of analytical understanding of fault characteristics and require some experience. In addition, previous location methods are more sensitive to line parameters, and the machine learning model obtained based on a specific line is not applicable to other lines, which restricts the application of the method. To solve the above problems, this paper proposes a double-ended combined fault location model based on Maximum Mean Discrepancy (MMD), which combined Convolutional Neural Network(CNN) and Long Short-Term Memory(LSTM). First, different transmission lines are categorized by MMD. Second, a double-ended CNN-LSTM combination model is built for similar lines, which autonomously extracts fault features in an end-to-end form, and then the weights of combination model are determined by the Q-learning algorithm. Finally, we obtain the fault distance prediction. Simulation studies show that the CNN-LSTM double-ended combined model based on MMD has good generalization performance for lines with different parameters, cracking the problem of specialized modeling of different lines while meeting the requirement of fault location accuracy.

Introduction: Cytochrome P450 (CYP) 3A4 is a major drug metabolizing enzyme for corticosteroids (CS). Epimedium has been used for asthma and variety of inflammatory conditions with or without CS. It is unknown whether epimedium has an effect on CYP 3A4 and how it interacts with CS. We sought to determine the effects of epimedium on CYP3A4 and whether it affects the anti-inflammatory function of CS and identify the active compound responsible for this effect.Methods: The effect of epimedium on CYP3A4 activity was evaluated using the Vivid CYP high-throughput screening kit. CYP3A4 mRNA expression was determined in human hepatocyte carcinoma (HepG2) cells with or without epimedium, dexamethasone, rifampin, and ketoconazole. TNF-α levels were determined following co-culture of epimedium with dexamethasone in a murine macrophage cell line (Raw 264.7). Active compound (s) derived from epimedium were tested on IL-8 and TNF-α production with or without corticosteroid, on CYP3A4 function and binding affinity.Results: Epimedium inhibited CYP3A4 activity in a dose-dependent manner. Dexamethasone enhanced the expression of CYP3A4 mRNA, while epimedium inhibited the expression of CYP3A4 mRNA and further suppressed dexamethasone enhancement of CYP3A4 mRNA expression in HepG2 cells (p < 0.05). Epimedium and dexamethasone synergistically suppressed TNF-α production by RAW cells (p < 0.001). Eleven epimedium compounds were screened by TCMSP. Among the compounds identified and tested only kaempferol significantly inhibited IL-8 production in a dose dependent manner without any cell cytotoxicity (p < 0.01). Kaempferol in combination with dexamethasone showed complete elimination of TNF-α production (p < 0.001). Furthermore, kaempferol showed a dose dependent inhibition of CYP3A4 activity. Computer docking analysis showed that kaempferol significantly inhibited the catalytic activity of CYP3A4 with a binding affinity of −44.73kJ/mol.Discussion: Inhibition of CYP3A4 function by epimedium and its active compound kaempferol leads to enhancement of CS anti-inflammatory effect.

Photovoltaic (PV) power generation is booming in rural areas, not only to meet the energy needs of local farmers but also to provide additional power to urban areas. Existing methods for estimating the spatial distribution of PV power generation potential either have low accuracy and rely on manual experience or are too costly to be applied in rural areas. In this paper, we discuss three aspects, namely, geographic potential, physical potential, and technical potential, and propose a large-scale and efficient PV potential estimation system applicable to rural rooftops in China. Combined with high-definition map images, we proposed an improved SegNeXt deep learning network to extract roof images. Using the national standard Design Code for Photovoltaic Power Plants (GB50797-2012) and the Bass model, computational results were derived. The average pixel accuracy of the improved SegNeXt was about 96%, which well solved the original problems of insufficient finely extracted edges, poor adhesion, and poor generalization ability and can cope with different types of buildings. Leizhou City has a geographic potential of 1500 kWh/m2, a physical potential of 25,186,181.7 m2, and a technological potential of 442.4 MW. For this paper, we innovatively used the Bass Demand Diffusion Model to estimate the installed capacity over the next 35 years and combined the Commodity Diffusion Model with the installed capacity, which achieved a good result and conformed to the dual-carbon “3060” plan for the future of China.

Reconfiguration is an important means to achieve optimal operation of the distribution network, but in actual distribution network operation, sectionalizing and tie switches are usually not switched frequently to adjust distribution network structure. In order to meet the practical application, this paper proposes a distribution network reconfiguration method based on comprehensive analysis of operation scenarios in order to find a reconfiguration solution with the best overall performance over a long term operating period. In this paper, different load combinations at each node of different time cross-sections in distribution network are regarded as distribution network operation scenarios. Firstly, the improved K-Means algorithm is used to cluster the operation scenarios of distribution network during the period to form typical operation scenarios, and the time share of each typical scenario in the entire time period is obtained. Then, the Time Interval Loss Index (TILI) is proposed to evaluate the total distribution network loss performance of reconfiguration solution in the entire time period, and then establish reconfiguration optimization model with this as optimization target, using genetic algorithm to optimize. Case study shows that the proposed method can adequately consider distribution network operation scenarios, and the overall network loss during time period is lower than the traditional reconfiguration method based on single operation scenario.

The effective assessment and prediction of the defect state of transmission lines can provide important technical support for the maintenance management of transmission lines. This paper proposes a method of correlation factors analysis and prediction for transmission line defect state based on association rule mining and RBF-SVM since the single operation parameter is often used in the analysis and prediction of transmission line defect state, and ignoring the influence of internal and external factors such as the meteorological conditions, operating conditions, etc. Firstly, according to the defect state assessment of transmission lines, based on the existing data, a characteristic library of the defect state and correlation factors is constructed by considering various relevant influencing factors. Then FP-Growth algorithm is introduced into the association rules mining, which can find the internal and external factors that have a strong association with defect, and the association rules can be used as the input feature of the prediction model, so as to avoid the influence of low association factors on the accuracy of defect state prediction. Finally, RBF-SVM was used to predict the defect state, and have a better prediction accuracy compared with three commonly used methods of the linear SVM, ANN and the decision tree. The proposed approach is illustrated by predicting the defect state of an overhead transmission line in a certain area. The results verify the effectiveness of the method and provide a certain reference for the maintenance of the transmission line.

Sensing the cloud movement information has always been a difficult problem in photovoltaic (PV) prediction. The information used by current PV prediction methods makes it challenging to accurately perceive cloud movements. The obstruction of the sun by clouds will lead to a significant decrease in actual PV power generation. The PV prediction network model cannot respond in time, resulting in a significant decrease in prediction accuracy. In order to overcome this problem, this paper develops a visual transformer model for PV prediction, in which the target PV sensor information and the surrounding PV sensor auxiliary information are used as input data. By using the auxiliary information of the surrounding PV sensors and the spatial location information, our model can sense the movement of the cloud in advance. The experimental results confirm the effectiveness and superiority of our model.

The future power grid will have more distributed energy sources, and the widespread access of distributed energy sources has the potential to improve the energy efficiency, resilience, and sustainability of the system. However, distributed energy, mainly wind power generation and photovoltaic power generation, has the characteristics of intermittency and strong randomness, which will bring challenges to the safe operation of the power grid. Accurate prediction of solar power generation with high spatial and temporal resolution is very important for the normal operation of the power grid. In order to improve the accuracy of distributed photovoltaic power generation prediction, this paper proposes a new distributed photovoltaic power generation prediction model: ROLL-GNN, which is defined as a prediction model based on rolling prediction of the graph neural network. The ROLL-GNN uses the perspective of graph signal processing to model distributed generation production timeseries data as signals on graphs. In the model, the similarity of data is used to capture their spatio-temporal dependencies to achieve improved prediction accuracy.

Cadmium (Cd) pollution poses a growing threat to rice production in acidic paddies. In south China, a common agricultural practice involves the combined utilization of Chinese milk vetch (M) and rice straw (R). However, it is unclear how the addition of lime to these amendments affects Cd bioavailability and accumulation in soil. Control (CK), chemical fertilizer (F), Chinese milk vetch + rice straw + chemical fertilizer (MRF), and Chinese milk vetch + rice straw + chemical fertilizer + lime (MRFL) treatments were applied to develop a kind of green, efficient, and practical amendment for acidic paddies. We conducted a microplot experiment to explore Cd immobilization in paddy soil and the Cd content in rice grains with these treatments. The results showed that compared with F, the rice Cd in the MRF and MRFL treatments were significantly decreased by 51.7% and 65.2% in early rice and 23.0% and 43.3% in late rice, respectively. Both the MRF and MRFL treatments significantly reduced soil available Cd and weak acid-extractable cadmium (Aci-Cd) concentrations and increased soil organic matter (SOM), exchangeable cation concentrations, and pH, which converted Cd into a stable form in soil. In addition, the MRF and MRFL treatments increased soil pH value by reducing soil exchangeable hydrogen ion concentration (E-H). Additionally, recombination of Cd forms was the primary factor in the reduction in available Cd concentration according to partial least squares path modeling (PLS-PM) analysis. The Cd concentration of rice grains was primarily associated with soil available Cd, soil pH value, and SOM. Overall, these results provide useful data and novel insights into reducing rice grain Cd in south China.

High precision short-term photovoltaic (PV) power prediction can reduce the damage associated with large-scale photovoltaic grid-connection to the power system. In this paper, a combination deep learning forecasting method based on variational mode decomposition (VMD), a fast correlation-based filter (FCBF) and bidirectional long short-term memory (BiLSTM) network is developed to minimize PV power forecasting error. In this model, VMD is used to extract the trend feature of PV power, then FCBF is adopted to select the optimal input-set to reduce the forecasting error caused by the redundant feature. Finally, the input-set is put into the BiLSTM network for training and testing. The performance of this model is tested by a case study using the public data-set provided by a PV station in Australia. Comparisons with common short-term PV power forecasting models are also presented. The results show that under the processing of trend feature extraction and feature selection, the proposed methodology provides a more stable and accurate forecasting effect than other forecasting models.

Multi-view subspace clustering has drawn significant attention in the pattern recognition and machine learning research community. However, most of the existing multi-view subspace clustering methods are still limited in two aspects. (1) The subspace representation yielded by the self-expression reconstruction model ignores the local structure information of the data. (2) The construction of subspace representation and clustering are used as two individual procedures, which ignores their interactions. To address these problems, we propose a novel multi-view subspace clustering method fusing local and global information for one-step multi-view clustering. Our contribution lies in three aspects. First, we merge the graph learning into the self-expression model to explore the local structure information for constructing the specific subspace representations of different views. Second, we consider the multi-view information fusion by integrating these specific subspace representations into one common subspace representation. Third, we combine the subspace representation learning, multi-view information fusion, and clustering into a joint optimization model to realize the one-step clustering. We also develop an effective optimization algorithm to solve the proposed method. Comprehensive experimental results on nine popular multi-view data sets confirm the effectiveness and superiority of the proposed method by comparing it with many state-of-the-art multi-view clustering methods.

Background: Alteration of commensal bacterial composition is associated with many inflammatory diseases. However, few studies have pinpointed the specific bacterial genes that may suppress host immune responses against microbes and maintain homeostasis in the host intestine. Methods: High-throughput screening was performed in Caenorhabditis elegans with a single gene knockout ut screening was performed in Caenorhabditis elegans with a single gene knockout Escherichia coli (E. coli) library and identified the immune suppression gene blc. The coding sequences of blc among different kinds of E. coli strains were aligned to identify the single nucleotide polymorphisms (SNPs). Physiological and biochemical experiments were performed in C. elegans and mice to explore the function of the blc variant. Findings: By screening 3983 E. coli mutants, we discovered that 9 bacterial genes, when deleted, activate innate immunity in the host C. elegans. Among these 9 genes, the gene encoding blc showed a distinctive SNP in many clinically pathogenic bacteria. We found that bacteria with this SNP, which converts Blc G84 to Blc E84, are highly enriched in the faeces of patients with inflammatory bowel disease (IBD). Exposure to BlcE84-encoding bacteria resulted in epithelial barrier disruption and immune activation in both worms and mice. Detailed analysis indicated that infection with BlcE84-encoding bacteria causes a significant decrease in LPE levels in the intestine and subsequently disrupts gut epithelial integrity in mice. Consistently, the levels of LPE in patients with IBD are significantly lower than those in healthy people. Finally, supplementation with LPE, which activates LPA1/PLCβ/PKC signaling, reversed the defects induced by BlcE84-encoding bacteria. Interpretation: Our results identified a novel bacterial gene, blc, in E. coli that regulates host gut integrity and immunity. Fund: The Ministry of Science and Technology of China; the National Natural Science Foundation of China; and the Natural Science Foundation of Jiangsu Province. Keywords: Bacterial lipocalin, Gut homeostasis, Epithelial barrier, lysophosphatidylethanolamine (LPE), Keio collection

Fault-cause identification plays a significant role in transmission line maintenance and fault disposal. With the increasing types of monitoring data, i.e., micrometeorology and geographic information, multiview learning can be used to realize the information fusion for better fault-cause identification. To reduce the redundant information of different types of monitoring data, in this paper, a hierarchical multiview feature selection (HMVFS) method is proposed to address the challenge of combining waveform and contextual fault features. To enhance the discriminant ability of the model, an ε-dragging technique is introduced to enlarge the boundary between different classes. To effectively select the useful feature subset, two regularization terms, namely l2,1-norm and Frobenius norm penalty, are adopted to conduct the hierarchical feature selection for multiview data. Subsequently, an iterative optimization algorithm is developed to solve our proposed method, and its convergence is theoretically proven. Waveform and contextual features are extracted from yield data and used to evaluate the proposed HMVFS. The experimental results demonstrate the effectiveness of the combined used of fault features and reveal the superior performance and application potential of HMVFS.