Search

Abstract Knowledge-aware recommendation systems often face challenges owing to sparse supervision signals and redundant entity relations, which can diminish the advantages of utilizing knowledge graphs for enhancing recommendation performance. To tackle these challenges, we propose a novel recommendation model named Dual-Intent-View Contrastive Learning network (DIVCL), inspired by recent advancements in contrastive and intent learning. DIVCL employs a dual-view representation learning approach using Graph Neural Networks (GNNs), consisting of two distinct views: a local view based on the user-item interaction graph and a global view based on the user-item-entity knowledge graph. To further enhance learning, a set of intents are integrated into each user-item interaction as a separate class of nodes, fulfilling three crucial roles in the GNN learning process: (1) providing fine-grained representations of user-item interaction features, (2) acting as evaluators for filtering relevant relations in the knowledge graph, and (3) participating in contrastive learning to strengthen the model’s ability to handle sparse signals and redundant relations. Experimental results on three benchmark datasets demonstrate that DIVCL outperforms state-of-the-art models, showcasing its superior performance. The implementation is available at: https://github.com/yzxx667/DIVCL .

Abstract MicroRNAs (miRNAs) are endogenous non-coding RNAs approximately 23 nucleotides in length, playing significant roles in various cellular processes. Numerous studies have shown that miRNAs are involved in the regulation of many human diseases. Accurate prediction of miRNA-disease associations is crucial for early diagnosis, treatment, and prognosis assessment of diseases. In this paper, we propose the Autoencoder Inductive Matrix Completion (AEIMC) model to identify potential miRNA-disease associations. The model captures interaction features from multiple similarity networks, including miRNA functional similarity, miRNA sequence similarity, disease semantic similarity, disease ontology similarity, and Gaussian interaction kernel similarity between miRNAs and diseases. Autoencoders are used to extract more complex and abstract data representations, which are then input into the inductive matrix completion model for association prediction. The effectiveness of the model is validated through cross-validation, stratified threshold evaluation, and case studies, while ablation experiments further confirm the necessity of introducing sequence and ontology similarities for the first time.

Machine learning techniques and computer-aided methods are now widely used in the pre-discovery tasks of drug discovery, effectively improving the efficiency of drug development and reducing the workload and cost. In this study, we used multi-source heterogeneous network information to build a network model, learn the network topology through multiple network diffusion algorithms, and obtain compressed low-dimensional feature vectors for predicting drug-target interactions (DTIs). We applied the metropolis-hasting random walk (MHRW) algorithm to improve the performance of the random walk with restart (RWR) algorithm, forming the basis by which the self-loop probability of the current node is removed. Additionally, the propagation efficiency of the MHRW was improved using the improved metropolis-hasting random walk (IMRWR) algorithm, facilitating network deep sampling. Finally, we proposed a correction of the transfer probability of the entire network after increasing the self-loop rate of isolated nodes to form the ISLRWR algorithm. Notably, the ISLRWR algorithm improved the area under the receiver operating characteristic curve (AUROC) by 7.53 and 5.72%, and the area under the precision-recall curve (AUPRC) by 5.95 and 4.19% compared to the RWR and MHRW algorithms, respectively, in predicting DTIs performance. Moreover, after excluding the interference of homologous proteins (popular drugs or targets may lead to inflated prediction results), the ISLRWR algorithm still showed a significant performance improvement.

Abstract The study of drug–target interaction plays an important role in the process of drug development. The subject of DTI forecasting has advanced significantly in the last several years, yielding numerous significant research findings and methodologies. Heterogeneous data sources provide richer information and comprehensive perspectives for drug–target interaction prediction, so many existing methods rely on heterogeneous networks, and graph embedding technology becomes an important technology to extract information from heterogeneous networks. These approaches, however, are less concerned with potential noisy information in heterogeneous networks and more focused on the extent of information extraction in those networks. Based on this, a potential DTI predictive network model called FBRWPC is proposed in this paper. It uses a fine-grained similarity selection program to first integrate similarity on similar networks and then a bidirectional random walk graph embedding learning method with restart to obtain an updated drug target interaction matrix. Through the use of similarity selection and fine-grained selection similarity integration, the framework can effectively filter out the noise present in heterogeneous networks and enhance the model's prediction performance. The experimental findings demonstrate that, even after being split up into four distinct types of data sets, FBRWPC can still retain great prediction performance, a sign of the model's resilience and good generalization.

Mapping water bodies from remotely sensed imagery is crucial for understanding hydrological and biogeochemical processes. The identification of water extent is mainly dependent on optical and synthetic aperture radar (SAR) images. However, the use of remote sensing for water body mapping is often undermined by the mixed pixel dilemma inherent to traditional hard classification approaches. At the same time, the presence of clouds in optical imagery and speckle noise in SAR imagery, coupled with the difficulty in differentiating between water-like surfaces and actual water bodies, significantly compromise the accuracy of water body identification. This paper proposes a DEEP feature collaborative convolutional neural network (CNN) for Water Super-Resolution Mapping based on Optical and SAR images (DeepOSWSRM), which collaboratively leverages Sentinel-1 and Sentinel-2 imagery to address the challenges of missing data and mixed pixels. The Sentinel-1 image provides complementary water distribution information for the cloudy areas of the Sentinel-2 image, while the Sentinel-2 image enhances the perception capabilities for small water bodies in the Sentinel-1 image. Using PlanetScope imagery as the true reference data, the effectiveness of the proposed method was assessed through two experimental scenarios: one utilizing synthetic coarse-resolution imagery degraded from Sentinel-1 and Sentinel-2 data and another using actual Sentinel-1 and Sentinel-2 data, encompassing both simulated and real cloud conditions. A comparative analysis was conducted against three state-of-the-art CNN-based water mapping methods and two CNN SRM methods. The findings demonstrate that the proposed DeepOSWSRM method successfully produces accurate, fine-resolution water body maps, with its performance mainly benefiting from the fusion of SAR and optical images.

This paper investigates the pure-cubic complex Ginzburg–Landau equation (PC-CGLE) with different nonlinearities such as Kerr law, power law and so on. We get the dynamic systems and show that solitons and periodic solutions exist through the complete discrimination system for the polynomial method (CDSPM). To verify these conclusions, we construct the traveling wave solution via the CDSPM, and some new solutions are also built. The soliton stability and modulation instability with two types of nonlinearities are discussed. Finally, by adding perturbed terms to the dynamic system, we obtain the largest Lyapunov exponents and the phase diagrams of the equation, which proves there are chaotic behaviors in PC-CGLE. The results such as Gaussian soliton solutions and chaotic behavior for PC-CGLE are initially discovered in the present paper.

Abstract Recently different cosmological measurements have shown a tension in the value of the Hubble constant, $$H_0.$$ H 0 . Assuming the $$\Lambda $$ Λ CDM model, the Planck satellite mission has inferred the Hubble constant from the cosmic microwave background (CMB) anisotropies to be $$H_0 = 67.4 \pm 0.5 \, \text {km} \, \text {s}^{-1} \, \text {Mpc}^{-1}.$$ H 0 = 67.4 ± 0.5 km s - 1 Mpc - 1 . On the other hand, low redshift measurements such as those using Cepheid variables and supernovae Type Ia (SNIa) have obtained a significantly larger value. For instance, Riess et al. reported $$H_0 = 73.04 \pm 1.04 \, \text {km} \, \text {s}^{-1} \, \text {Mpc}^{-1},$$ H 0 = 73.04 ± 1.04 km s - 1 Mpc - 1 , which is $$5\sigma $$ 5 σ apart of the prediction from Planck observations. This tension is a major problem in cosmology nowadays, and it is not clear yet if it comes from systematic effects or new physics. The use of new methods to infer the Hubble constant is therefore essential to shed light on this matter. In this paper, we discuss using the ages of the oldest astrophysical objects (OAO) to probe the Hubble tension. We show that, although this data can provide additional information, the method can also artificially introduce a tension. Reanalyzing the ages of 114 OAO, we obtain that the constraint in the Hubble constant goes from slightly disfavoring local measurements to favoring them.

Crop mapping is vital in ensuring food production security and informing governmental decision-making. The satellite-normalized difference vegetation index (NDVI) obtained during periods of vigorous crop growth is important for crop species identification. Sentinel-2 images with spatial resolutions of 10, 20, and 60 m are widely used in crop mapping. However, the images obtained during periods of vigorous crop growth are often covered by clouds. In contrast, time-series moderate-resolution imaging spectrometer (MODIS) images can usually capture crop phenology but with coarse resolution. Therefore, a time-series-based spatiotemporal fusion network (TSSTFN) was designed to generate TSSTFN-NDVI during critical phenological periods for finer-scale crop mapping. This network leverages multi-temporal MODIS-Sentinel-2 NDVI pairs from previous years as a reference to enhance the precision of crop mapping. The long short-term memory module was used to acquire data about the time-series change pattern to achieve this. The UNet structure was employed to manage the spatial mapping relationship between MODIS and Sentinel-2 images. The time distribution of the image sequences in different years was inconsistent, and time alignment strategies were used to process the reference data. The results demonstrate that incorporating the predicted critical phenological period NDVI consistently yields better crop classification performance. Moreover, the predicted NDVI trained with time-consistent data achieved a higher classification accuracy than the predicted NDVI trained with the original NDVI.

Prediction of drug-target interactions (DTIs) plays an important role in drug development. However, traditional laboratory methods to determine DTIs require a lot of time and capital costs. In recent years, many studies have shown that using machine learning methods to predict DTIs can speed up the drug development process and reduce capital costs. An excellent DTI prediction method should have both high prediction accuracy and low computational cost. In this study, we noticed that the previous research based on deep forests used XGBoost as the estimator in the cascade, we applied LightGBM instead of XGBoost to the cascade forest as the estimator, then the estimator group was determined experimentally as three LightGBMs and three ExtraTrees, this new model is called LGBMDF. We conducted 5-fold cross-validation on LGBMDF and other state-of-the-art methods using the same dataset, and compared their Sn, Sp, MCC, AUC and AUPR. Finally, we found that our method has better performance and faster calculation speed.

Male infertility has always been one of the important factors affecting the infertility of couples of gestational age. The reasons that affect male infertility includes living habits, hereditary factors, etc. Identifying the genetic causes of male infertility can help us understand the biology of male infertility, as well as the diagnosis of genetic testing and the determination of clinical treatment options. While current research has made significant progress in the genes that cause sperm defects in men, genetic studies of sperm content defects are still lacking. This article is based on a dataset of gene expression data on the X chromosome in patients with azoospermia, mild and severe oligospermia. Due to the difference in the degree of disease between patients and the possible difference in genetic causes, common classical clustering methods such as k-means, hierarchical clustering, etc. cannot effectively identify samples (realize simultaneous clustering of samples and features). In this paper, we use machine learning and various statistical methods such as hypergeometric distribution, Gibbs sampling, Fisher test, etc. and genes the interaction network for cluster analysis of gene expression data of male infertility patients has certain advantages compared with existing methods. The cluster results were identified by differential co-expression analysis of gene expression data in male infertility patients, and the model recognition clusters were analyzed by multiple gene enrichment methods, showing different degrees of enrichment in various enzyme activities, cancer, virus-related, ATP and ADP production, and other pathways. At the same time, as this paper is an unsupervised analysis of genetic factors of male infertility patients, we constructed a simulated data set, in which the clustering results have been determined, which can be used to measure the effect of discriminant model recognition. Through comparison, it finds that the proposed model has a better identification effect.

The incidence of breast cancer in women has surpassed that of lung cancer as the world’s leading new cancer case. Regular screening and measures become an effective way to prevent breast cancer and also provide a good foundation for later treatment. Women should receive regular checkups in the hospital after reaching a certain age. The use of computer-aided technology can improve the accuracy and efficiency of physicians’ decision-making. Data pre-processing is required before data analysis, and 16 features are selected using a correlation-based feature selection method. In this paper, meta-learning and Artificial Neural Networks (ANN) are combined to create a hybrid algorithm. The proposed hybrid algorithm for predicting breast cancer was attempted to achieve 98.74% accuracy and 98.02% F1-score by creating a combination of various meta-learning models whose output was used as input features for creating ANN models. Therefore, the hybrid algorithm proposed in this paper can obtain better prediction results than a single model.

In this study, the geometric structure, strain energy, and electronic properties of armchair C2O nanotubes (A_C2ONTs(n,n)) and zigzag C2O nanotubes (Z_C2ONTs(n,0)) are studied in detail. The results show that these nanotubes behave as special shapes; the section of Z_C2ONTs(n,0) along the axial direction is an n-edge type, and the section tends to be round with an increase in n while the boundary of section along the axial direction for A_C2ONTs(n,n) fluctuates. With an increase in n, the fluctuation disappears gradually, and the section also tends to be round. Compared with C2O nanosheets (g_C2O), C2O nanotubes have similar or even smaller strain energy, when the tube diameter is greater than or equal to 15 Å, the strain energy begins to show a negative value, and the negative value tends to be stable as the pipe diameter increases. Z_C2ONTs start to show negative strain energies from n = 8 and A_C2ONTs from n = 12, indicating their higher stabilities relative to the g_C2O sheets. The calculation of the electronic band structure shows that Z_C2ONTs are an indirect band gap semiconductor, and A_C2ONT is a direct band gap semiconductor. The band gap value of Z_C2ONTs first increases and then stabilizes with the diameters of the nanotubes, and the band gap value of A_C2ONTs decreases with the increase in the nanotubes and tends to the band gap value of the g_C2O sheet. In addition, the electronic properties and thermoelectric properties of C2ONTs(n=4,6,8) before and after N-doping were also studied. We found that N-doping changed the electronic and thermoelectric properties of C2ONTs. It reduced the nanotube band gap value and significantly improved the thermoelectric figures of merit of Z_C2ONTs(n=4,6,8) and A_C2ONT(4,4), which comprised an effective strategy to improve the thermoelectric figure of merit of nanotubes. The results showed that the C2ONTs had potential as thermoelectric materials after N-doping, which provided important guidance for designing low-dimensional g_C2O nanostructures.

Switching circuits provide an interesting strategy for modularization construction of digital computing. Multi-step reactions on DNA origami template are essentially the same as switching circuits. Herein, DNA switch circuits(DSCs) based on binary tree is used to solve arbitrary digital computing. The establishment and implementation of DSCs is completed on DNA origami template. The nodes in the binary tree are mapped to hairpins on DNA origami template and the switches are mapped to the hairpins which is abundant in the solution. DSCs operate on the basis of a hybridization chain reaction on DNA origami substrates. The results are read using bulk fluorescence spectroscopic methods. Compared with multi input logic “AND” gate, the number of DNA strands (anchored hairpins and fuels) required by DSCs is reduced more than 33.33% and 66.67%, respectively. The DSCs are used to construct the full-subtractor and prime number discriminator, which indicates this strategy can construct arbitrary digital circuits.

Monitoring the spatio-temporal dynamic changes of reservoir surface water area (RSWA) values is crucial when investigating global water resource reserves. Moderate Resolution Imaging Spectroradiometer (MODIS) remotely sensed imagery has been extensively used for this purpose, but the coarse spatial resolution limits the accuracy when monitoring the small water area variations of reservoirs. In this paper, we introduce a novel approach for the monitoring of RSWA variation by integrating MODIS and Landsat imagery to overcome this problem. In the proposed method, the sub-pixel water fraction information in the MODIS images is explored based on the C/M ratio model, where C/M represents the ratio between the reflectance of a land pixel and that of a water pixel. By using the series of RSWA values extracted from Landsat images as the reference, the C/M values extracted from MODIS images are used to construct regression functions with the overall RSWA for each MODIS pixel. An RSWA time series is then predicted from the extracted daily C/M values with the use of the regression functions constructed in the study. In a case study of Dale Hollow Reservoir in the United States, the performance of the proposed method was assessed, and the resulting Pearson's correlation coefficient between the predicted RSWA and water level (WL) reached 0.88, which was higher than that of the conventional pixel analysis based method and the spectral unmixing based sub-pixel analysis method, confirming the effectiveness of the proposed method.

Cardiotocography (CTG) recorded fetal heart rate and its temporal relationship with uterine contractions. CTG intelligent classification plays an important role in evaluating fetal health and protecting fetal normal growth and development throughout pregnancy. At the feature selection level, this study uses the Apriori algorithm to search frequent item sets for feature extraction. At the level of the classification model, the combination model of AdaBoost and random forest with the highest classification accuracy is finally selected by comparing various models. The suspicious class data in the CTG data set affect the overall classification accuracy. The number of suspicious class data is predicted by the multi-model ensemble method. Finally, the data set is fused from three classifications to two classifications. The classification accuracy is 0.976, and the AUC is 0.98, which significantly improves the classification effect. In conclusion, the method used in this study has high accuracy in model classification, which is helpful to improve the accuracy of fetal abnormality detection.

Numerous pieces of evidence have indicated that microRNA (miRNA) plays a crucial role in a series of significant biological processes and is closely related to complex disease. However, the traditional biological experimental methods used to verify disease-related miRNAs are inefficient and expensive. Thus, it is necessary to design some excellent approaches to improve efficiency. In this work, a novel method (CFSAEMDA) is proposed for the prediction of unknown miRNA–disease associations (MDAs). Specifically, we first capture the interactive features of miRNA and disease by integrating multi-source information. Then, the stacked autoencoder is applied for obtaining the underlying feature representation. Finally, the modified cascade forest model is employed to complete the final prediction. The experimental results present that the AUC value obtained by our method is 97.67%. The performance of CFSAEMDA is superior to several of the latest methods. In addition, case studies conducted on lung neoplasms, breast neoplasms and hepatocellular carcinoma further show that the CFSAEMDA method may be regarded as a utility approach to infer unknown disease–miRNA relationships.

This paper addresses the adaptive asymptotic tracking control problem for nonlinear systems whose virtual control gains are unknown nonlinear functions of system states. Only in the first step, the Nussbaum gain technique is utilized to handle the uncertain virtual control gain. In the remaining steps, virtual control gains are dealt with by constructing novel control laws without the approximation of the uncertain nonlinear functions and external disturbances by neural networks or fuzzy logic. New adaptive laws are defined to compensate for unknown virtual control gains, uncertain parameters, and external disturbances. Finally, an adaptive tracking controller is designed and applied to the control of a 3-order robot system, which guarantees the boundedness of all the signals in the closed-loop system and asymptotic stability of the tracking error.

The image encryption schemes combining chaotic maps, DNA coding, and DNA sequence operation can effectively protect the image. In this paper, a double-layer image encryption scheme is proposed by combining chaotic maps with DNA strand displacement (DSD). Chaotic maps are used to generate pseudorandom sequences and perform routine scrambling and diffusion operations on the plaintext image. We propose three DSD-based encryption rules according to the diversity of DNA strand displacement, and these three encryption rules are used to encrypt the image at the DNA sequence level. The plaintext image can be transformed into the cipher image, which is difficult to be recognized without the correct keys through the double-layer encryption at the level of chaotic maps and DNA. Simulation results and security analysis show that the proposed encryption scheme can effectively protect image information and resist conventional information attacks.

DNA logic gates are an important branch of DNA computing and have a wide range of applications in DNA computing. In this study, logic circuits of AND gate and NAND gate are built on origami substrate. The realization of AND gate uses polymerase strand displacement (PSD) reaction and hybridization chain reaction (HCR). If there is a fluorescent band “1” displayed, the result is true. The realization of the NAND gate requires a cyclic reaction. If there is a fluorescent band “A” or “T” displayed, the result is true; if no fluorescent band is displayed, the result is false.

Optical remote sensing imagery is commonly used to monitor the spatial and temporal distribution patterns of inland waters. Its usage, however, is limited by cloud contamination, which results in low-quality images or missing values. Selecting cloud-free scenes or combining multi-temporal images to produce a cloud-free composite image can partially overcome this problem at the cost of the monitoring frequency. Predicting the spectral values of cloudy areas based on the spectral characteristics is a possible solution; however, this is not appropriate for water because it changes rapidly. Reconstructing cloud-covered water areas using historical water-distribution data has good performance, but such methods are typically only suitable for lakes and reservoirs, not over vast and complex terrain. This paper proposes a category-based approach to reconstruct the water distribution in cloud-contaminated images using a spatiotemporal dependence model. The proposed method predicts the class label (water or land) of a cloudy pixel based on the neighboring pixel labels and those at the same position in images acquired on other dates according to historical spatiotemporal water-distribution data. The method was evaluated through eight experiments in different study regions using Landsat and Sentinel-2 images. The results demonstrated that the proposed method could yield high-quality cloud-free classification maps and provide good water-extraction accuracy and consistency in most hydrological conditions, with an overall accuracy of up to 98%. The accuracy and practicality of the method render it promising for applications across a wide range of future research and monitoring efforts.

The estimation of the large and high-dimensional covariance matrix and precision matrix is a fundamental problem in modern multivariate analysis. It has been widely applied in economics, finance, biology, social networks and health sciences. However, the traditional sample estimators perform poorly for large and high-dimensional data. There are many approaches to improve the covariance matrix estimation. The large dynamic conditional correlation model based on the nonlinear shrinkage and its application in portfolio selection attract increasing attention. In the estimation of the unconditional covariance matrix, the graphical lasso is more robust than the nonlinear shrinkage model, and the leptokurtic and fat tail characteristics of the asset returns are also more obvious. This article proposes improved large dynamic covariance matrix estimation based on the graphical lasso models under the multivariate normal distribution (glasso) and t distribution (tlasso), and the corresponding dynamic conditional correlation glasso and tlasso approaches are developed. To verify the effectiveness and robustness of the proposed methods, we conduct simulations and then apply the models to the classic Markowitz portfolio selection problem. Simulations and empirical results show that the combined dynamic conditional correlation glasso and tlasso approaches outperform the current dynamic covariance matrix estimators.

The estimation of the covariance matrix is an important topic in the field of multivariate statistical analysis. In this paper, we propose a new estimator, which is a convex combination of the linear shrinkage estimation and the rotation-invariant estimator under the Frobenius norm. We first obtain the optimal parameters by using grid search and cross-validation, and then, we use these optimal parameters to demonstrate the effectiveness and robustness of the proposed estimation in the numerical simulations. Finally, in empirical research, we apply the covariance matrix estimation to the portfolio optimization. Compared to the existing estimators, we show that the proposed estimator has better performance and lower out-of-sample risk in portfolio optimization.

In recent years, single-cell RNA sequencing technology (scRNA-seq) has developed rapidly and has been widely used in biological and medical research, such as in expression heterogeneity and transcriptome dynamics of single cells. The investigation of RNA velocity is a new topic in the study of cellular dynamics using single-cell RNA sequencing data. It can recover directional dynamic information from single-cell transcriptomics by linking measurements to the underlying dynamics of gene expression. Predicting the RNA velocity vector of each cell based on its gene expression data and formulating RNA velocity prediction as a classification problem is a new research direction. In this paper, we develop a cascade forest model to predict RNA velocity. Compared with other popular ensemble classifiers, such as XGBoost, RandomForest, LightGBM, NGBoost, and TabNet, it performs better in predicting RNA velocity. This paper provides guidance for researchers in selecting and applying appropriate classification tools in their analytical work and suggests some possible directions for future improvement of classification tools.

Because of the abundance and low cost of sodium, sodium-ion batteries (SIBs) are next-generation energy storage mediums. Furthermore, SIBs have become an alternative option for large-scale energy storage systems. Because the electrolyte is a critical component of SIBs, fluorination is performed to improve the cycling performance of electrolytes. Based on the first-principles study, we investigated the effects of the type, quantity, and relative position relationships of three fluorinated units, namely -CF1, -CF2, and -CF3, on the cyclic ester molecule ethylene carbonate (EC) and the linear ether molecule 1,2-dimethoxylethane (DME). The optimal fluorination was proposed for EC and DME by studying the bond length, highest occupied molecular orbital, lowest unoccupied lowest orbital, and other relevant parameters. The results revealed that for EC, the optimal fluorination is 4 F fluorination based on four -CF1 units; for DME, CF3CF1CF1-, CF3CF2CF2-, CF3CF1CF2CF3, and CF3CF2CF2CF3, four combinations of three -CF1, -CF2, and -CF3 units are optimal. The designed fluorinated EC and DME exhibited a wide electrochemical stability window and high ionic solvation ability, which overcomes the drawback of conventional solvents and can improve SIB cycling performance.

To study the influence of earthquakes and engineering disturbances on the deformation of deeply buried rock masses, shear tests were carried out on anchored sandstone rock masses, anchored marble rock masses, and anchored granite rock masses under creep fatigue loading, and a new creep fatigue model was established to characterize the deformation characteristics of anchored rock masses under creep fatigue loading. The creep fatigue curves of different lithologies clearly show three stages: creep attenuation, steady-state creep, and accelerated creep. Fatigue loading can increase the creep of anchored specimens, and the lower the rock strength is, the higher the creep variable under fatigue loading is. However, for the same rock strength, with the increase in load level, the creep variable produced by creep fatigue load presents a linear downward trend. Considering the changes in the mechanical properties of the anchored rock mass under creep fatigue loading, the creep fatigue model of anchored rock masses is established by introducing a function of the fatigue shear modulus, and the accuracy and applicability of the model are verified by laboratory creep fatigue test data. The model provides a theoretical basis for the study of anchored rock mass support under low-frequency earthquakes or blasting loads.

Long noncoding RNAs (lncRNAs) are a class of RNAs longer than 200 nt and cannot encode the protein. Studies have shown that lncRNAs can regulate gene expression at the epigenetic, transcriptional, and posttranscriptional levels, which are not only closely related to the occurrence, development, and prevention of human diseases, but also can regulate plant flowering and participate in plant abiotic stress responses such as drought and salt. Therefore, how to accurately and efficiently identify lncRNAs is still an essential job of relevant researches. There have been a large number of identification tools based on machine-learning and deep learning algorithms, mostly using human and mouse gene sequences as training sets, seldom plants, and only using one or one class of feature selection methods after feature extraction. We developed an identification model containing dicot, monocot, algae, moss, and fern. After comparing 20 feature selection methods (seven filter and thirteen wrapper methods) combined with seven classifiers, respectively, considering the correlation between features and model redundancy at the same time, we found that the WOA-XGBoost-based model had better performance with 91.55%, 96.78%, and 91.68% of accuracy, AUC, and F1_score. Meanwhile, the number of elements in the feature subset was reduced to 23, which effectively improved the prediction accuracy and modeling efficiency.

Nanostructures with information processing play an important role in many fields. It is an excellent approach to the application that DNA nanostructures represented by DNA origami molecules combine with the hybrid chain reaction. In this paper, the assignment problem is mapped to a combinatorial graph on the DNA origami substrate. The graph has several modules corresponding to the time efficiency matrix of the assignment problem. The starting chain of the corresponding module is hybridized with the hairpin structure of the starting point, and the corresponding module is opened to emit light. The feasible solution to the problem can be obtained by observing the light-emitting fluorescent numbers of the opened modules. The fluorescent numbers of all the opened modules are added up on the same origami substrate, then different opening methods in different test tubes are compared, and the optimal solution is obtained.

The noise2noise-based despeckling method, capable of training the despeckling deep neural network with only noisy synthetic aperture radar (SAR) image, has presented very good performance in recent research. This method requires a fine-registered multi-temporal dataset with minor time variance and uses similarity estimation to compensate for the time variance. However, constructing such a training dataset is very time-consuming and may not be viable for a certain practitioner. In this article, we propose a novel single-image-capable speckling method that combines the similarity-based block-matching and noise referenced deep learning network. The denoising network designed for this method is an encoder–decoder convolutional neural network and is accommodated to small image patches. This method firstly constructs a large number of noisy pairs as training input by similarity-based block-matching in either one noisy SAR image or multiple images. Then, the method trains the network in a Siamese manner with two parameter-sharing branches. The proposed method demonstrates favorable despeckling performance with both simulated and real SAR data with respect to other state-of-the-art reference filters. It also presents satisfying generalization capability as the trained network can despeckle well the unseen image of the same sensor. The main advantage of the proposed method is its application flexibility. It could be trained with either one noisy image or multiple images. Furthermore, the despeckling could be inferred by either the ad hoc trained network or a pre-trained one of the same sensor.

To investigate the engineering feasibility of translucent lightweight aggregate concrete (TLAC) in sustainable buildings, this paper focuses on testing the light transmission properties and mechanical performance of TLACs, the light transmission effect in building interiors is simulated and evaluated using automated steps seamlessly integrated with REVIT software and Radiance software. The preparation process of TLACs and the mixture proportions of raw materials were designed by combining translucent concrete technology with self-compacting mortar. The results indicate that the light transmission of TLAC increases with the volume of polymethylmethacrylate optical fiber (POF) incorporated and the diameter of POF, while the compressive strength tends to increase first and then decrease with the increasing volume fraction of POF in the TLAC samples. Moreover, the building lighting simulations show that indoor illuminance will improve significantly with the use of TLACs, which further illustrates the energy savings and structural performance improvements owing to the application of TLAC as a modern construction material.

A long-span sports centre generally comprises multiple stadiums and gymnasiums, for which mutual interference effects of wind-induced snow motion are not explicitly included in the specifications of various countries. This problem is addressed herein by performing wind tunnel tests and numerical simulations to investigate the snow distribution and mutual interference effect on the roofs of long-span stadiums and gymnasiums. The wind tunnel tests were used to analyse the influences of the opening direction (0°, 90°, 180°, and 270°) and spacing (0.3 L, 0.5 L, 1 L, 1.5 L, 2 L, and 2.5 L, where L is the gymnasium span) of the stadium and gymnasium. The wind tunnel tests and numerical simulations were used to analyse the influence of the wind direction angle (from 0° to 315°, there are a total of eight groups in 45° intervals). The following results were obtained. The stadium opening had a significant effect on the snow distribution on the surface of the two structures. An even snow distribution was obtained when the stadium opened directly facing the gymnasium, which corresponded to the safest condition for the structures’ surfaces. As the spacing between the buildings increased, the interference effect between the two structures was reduced. The interference was negligible for a spacing of 2 L. The stadium had the most significant amplification interference effect on the gymnasium for a wind direction angle of 45°, which was extremely unfavourable to the safety of the structure. The most favourable wind direction angle was 270°, where there were both amplification interference and blockage interference.

DNA origami is the application of self-assembly in nanotechnology. The combination of DNA origami and hybridization chain reaction is one of the important application methods of DNA origami. In this paper, DNA origami is used to design the cipher pattern on the base of origami. The cipher chain, which is put into the reaction solution, hybridizes with the molecular beacon and the hairpin structure that form the cipher pattern to build a DNA origami model that can decode the pattern. The cipher chain consists of the starting chain and the intermediate chain. When the cipher is correct, the cipher chain can open the molecular beacon and the hairpin structure to display the cipher pattern on the origami base in the solution.

The outbreak of coronavirus disease 2019 (COVID-19) has caused a global disaster, seriously endangering human health and the stability of social order. The purpose of this study is to construct a nonlinear combinational dynamic transmission rate model with automatic selection based on forecasting effective measure (FEM) and support vector regression (SVR) to overcome the shortcomings of the difficulty in accurately estimating the basic infection number R0 and the low accuracy of single model predictions. We apply the model to analyze and predict the COVID-19 outbreak in different countries. First, the discrete values of the dynamic transmission rate are calculated. Second, the prediction abilities of all single models are comprehensively considered, and the best sliding window period is derived. Then, based on FEM, the optimal sub-model is selected, and the prediction results are nonlinearly combined. Finally, a nonlinear combinational dynamic transmission rate model is developed to analyze and predict the COVID-19 epidemic in the United States, Canada, Germany, Italy, France, Spain, South Korea, and Iran in the global pandemic. The experimental results show an the out-of-sample forecasting average error rate lower than 10.07% was achieved by our model, the prediction of COVID-19 epidemic inflection points in most countries shows good agreement with the real data. In addition, our model has good anti-noise ability and stability when dealing with data fluctuations.

The nonuniform distribution of snow around structures with holes is extremely unfavorable for structural safety, and the mechanism of wind-snow interaction between adjacent structures with holes needs to be explored. Therefore, a wind tunnel simulation was performed, in which quartz particles with an average particle size of 0.14 mm as snow particles were used, and cubes with dimensions of 100 mm × 100 mm × 100 mm each containing a hole with the size of 20 mm × 20 mm were employed as structures. Firstly, the quality of a small low-speed wind tunnel flow field was tested, and then the effects of hole orientation (hole located on the windward side, leeward side, and other vertical sides) and absence of holes on the surface of a single cube were studied. Furthermore, the effects of different hole locations (respectant position, opposite position, and dislocation) and relative spacing (50 mm, 100 mm, and 150 mm) on the surfaces of two cubes and the snow distribution around them were investigated. It was concluded that the presence and location of hole had a great influence on snow distribution around cubes. Snow distribution was favorable when hole was located on the other vertical sides of the test specimen. The most unfavorable snow distribution was obtained when the holes on the two-holed sides of the cubes were respectant with a maximum snow depth coefficient of 1.4. A significant difference was observed in the snow depths of two sides of cubes when holes were dislocated. When two holes were respectant, surrounding snow depth was decreased, and the maximum snow depth on model surface area was increased with the increase of spacing. Wind tunnel tests on holed cubes provided a reference for the prediction of snow load distribution of typical structures with holes.

Geostationary satellite land surface temperature (GLST) data are important for various dynamic environmental and natural resource applications for terrestrial ecosystems. Due to clouds, shadows, and other atmospheric conditions, the derived LSTs are often missing a large number of values. Reconstructing the missing values is essential for improving the usability of the geostationary satellite LST data. However, current reconstruction methods mainly aim to fill the values of a small number of invalid pixels with many valid pixels, which can provide useful land surface temperature values. When the missing data extent becomes large, the reconstruction effect will worsen because the relationship between different spatiotemporal geostationary satellite LSTs is complex and highly nonlinear. Inspired by the superiority of the deep convolutional neural network (CNN) in solving highly nonlinear and dynamic problems, a multiscale feature connection CNN model is proposed to fill missing LSTs with large missing regions. The proposed method has been tested on both FengYun-2G and Meteosat Second Generation-Spinning Enhanced Visible and InfraRed Imager geostationary satellite LST datasets. The results of simulated and actual experiments show that the proposed method is accurate to within about 1 °C, with 70% missing data rates. This is feasible and effective for large regions of LST reconstruction tasks.