Search

Interleukin-6 (IL-6) is a potent glycoprotein that plays a crucial role in regulating innate and adaptive immunity, as well as metabolism. The expression and release of IL-6 are closely correlated with the severity of various diseases. IL-6-inducing peptides are critical for the development of immunotherapy and diagnostic biomarkers for some diseases. Most existing methods for predicting IL-6-induced peptides use traditional machine learning methods, whose feature selection is based on prior knowledge. In addition, none of these methods take into account the three-dimensional (3D) structure of peptides, which is essential for their functional properties. In this study, we propose a novel IL-6-inducing peptide prediction method called DGIL-6, which integrates 3D structural information with graph neural networks. DGIL-6 represents a peptide sequence as a graph, where each amino acid is treated as a node, and the adjacency matrix, representing the relationships between nodes, is derived from the predicted residue contact graph of the peptide sequence. In addition to commonly used amino acid representations, such as one-hot encoding and position encoding, the pre-trained model ESM-1b is employed to extract amino acid features as node features. In order to simultaneously consider node weights and information updates, a dual-channel method combining Graph Attention Network (GAT) and Graph Convolutional Network (GCN) is adopted. Finally, the extracted features from both channels are merged for the classification of IL-6-inducing peptides. A series of experiments including cross-validation, independent testing, ablation studies, and visualizations demonstrate the effectiveness of the DGIL-6 method.

Blackleg disease, a major threat to Brassica crops worldwide, is primarily caused by the pathogen Leptosphaeria biglobosa. Investigating the genetic variation of L. biglobosa is crucial for managing and preventing the disease in Brassica napus. To date, there is scarce genomic variation information available for populations of L. biglobosa in China. In this study, 73 L. biglobosa strains of canola stalks were collected from 12 provinces in China and subjected to re-sequencing. The 73 assemblies averaged 1340 contigs, 72,123 bp N50, and 30.17 Mb in size. In total, 9409 core orthogroups and 867 accessory orthogroups were identified. A total of 727,724 mutation loci were identified, including 695,230 SNPs and 32,494 indels. Principal component analysis (PCA) and population structure analysis showed that these strains could be divided into seven subgroups. The strains in most provinces were clustered into a single subgroup, suggesting a strong influence of the geographic environment on strain variation. The average nucleotide diversity (θπ) of all strains was 1.03 × 10−3, indicating important genetic diversity among strains from different regions of China. This study provides valuable resources for future comparative genomics, gives new insights into the population evolution of L. biglobosa, and supports the development of strategies for managing blackleg disease in canola.

To explore the application of a traditional machine learning model in the intelligent management of pigs, in this paper, the influence of PCA pre-treatment on pig face identification with RF is studied. By this testing method, the parameters of two testing schemes, one adopting RF alone and the other adopting RF + PCA, were determined to be 65 and 70, respectively. With individual identification tests carried out on 10 pigs, accuracy, recall, and f1-score were increased by 2.66, 2.76, and 2.81 percentage points, respectively. Except for the slight increase in training time, the test time was reduced to 75% of the old scheme, and the efficiency of the optimized scheme was greatly improved. It indicates that PCA pre-treatment positively improved the efficiency of individual pig identification with RF. Furthermore, it provides experimental support for the mobile terminals and the embedded application of RF classifiers.

Intelligent detection is vital for achieving the intelligent picking operation of daylily, but complex field environments pose challenges due to branch occlusion, overlapping plants, and uneven lighting. To address these challenges, this study selected an intelligent detection model based on YOLOv5s for daylily, the depth and width parameters of the YOLOv5s network were optimized, with Ghost, Transformer, and MobileNetv3 lightweight networks used to optimize the CSPDarknet backbone network of YOLOv5s, continuously improving the model’s performance. The experimental results show that the original YOLOv5s model increased mean average precision (mAP) by 49%, 44%, and 24.9% compared to YOLOv4, SSD, and Faster R-CNN models, optimizing the depth and width parameters of the network increased the mAP of the original YOLOv5s model by 7.7%, and the YOLOv5s model with Transformer as the backbone network increased the mAP by 0.2% and the inference speed by 69% compared to the model after network parameter optimization. The optimized YOLOv5s model provided precision, recall rate, mAP, and inference speed of 81.4%, 74.4%, 78.1%, and 93 frames per second (FPS), which can achieve accurate and fast detection of daylily in complex field environments. The research results can provide data and experimental references for developing intelligent picking equipment for daylily.

The worldwide pandemic of COVID-19 has become a global public health crisis. Various clinical diagnosis methods have been developed to distinguish COVID-19–infected patients from healthy people. The nucleic acid test is the golden standard for virus detection as it is suitable for early diagnosis. However, due to the low amount of viral nucleic acid in the respiratory tract, the sensitivity of nucleic acid detection is unsatisfactory. As a result, serological screening began to be widely used with the merits of simple procedures, lower cost, and shorter detection time. Serological tests currently include the enzyme-linked immunosorbent assay (ELISA), lateral flow immunoassay (LFIA), and chemiluminescence immunoassay (CLIA). This review describes various serological methods, discusses the performance and diagnostic effects of different methods, and points out the problems and the direction of optimization, to improve the efficiency of clinical diagnosis. These increasingly sophisticated and diverse serological diagnostic technologies will help human beings to control the spread of COVID-19.

This study presents a machine learning-based method for the destructive power assessment of earthquake to structures. First, the analysis procedure of the method is presented, and the backpropagation neural network (BPNN) and convolutional neural network (CNN) are used as the machine learning algorithms. Second, the optimized BPNN architecture is obtained by discussing the influence of a different number of hidden layers and nodes. Third, the CNN architecture is proposed based on several classical deep learning networks. To build the machine learning models, 50,570 time-history analysis results of a structural system subjected to different ground motions are used as training, validation, and test samples. The results of the BPNN indicate that the features extraction method based on the short-time Fourier transform (STFT) can well reflect the frequency-/time-domain characteristics of ground motions. The results of the CNN indicate that the CNN exhibits better accuracy (R2 = 0.8737) compared with that of the BPNN (R2 = 0.6784). Furthermore, the CNN model exhibits remarkable computational efficiency, the prediction of 1000 structures based on the CNN model takes 0.762 s, while 507.81 s are required for the conventional time-history analysis (THA)-based simulation. Feature visualization of different layers of the CNN reveals that the shallow to deep layers of the CNN can extract the high to low-frequency features of ground motions. The proposed method can assist in the fast prediction of engineering demand parameters of large-number structures, which facilitates the damage or loss assessments of regional structures for timely emergency response and disaster relief after earthquake.

Buildings in coastal cities are susceptible to chloride ion attack and the seismic performance of these buildings can be impaired due to corrosion of reinforcements. In this study, a regional seismic damage simulation method that considers the influence of corrosion-induced seismic performance degradation is proposed. Firstly, the framework of the method is introduced, and the simulation process is presented. Secondly, experimental data of corroded reinforced concrete (RC) components are collected to obtain the reduction rules of component level backbone curve parameters (i.e., initial stiffness, peak carrying capacity, peak displacement, and ultimate carrying capacity). Afterwards, pushover analyses of typical RC frames in different corrosion conditions (i.e., degree of corrosion of components and proportion of corroded components) are conducted to acquire the reduction rules of interstory backbone curve parameters of corroded RC frame structures. Finally, RC frame structures, in Shenzhen city, are simulated using different corrosion scenarios. Simulated results indicate that some buildings along the coastline are affected by airborne chloride-induced corrosion and severe seismic damage can be observed. Moreover, some buildings that are far from the coastline can also experience severe seismic damage due to irregular use of sea sand as constructional material. The proposed method can be used to simulate the seismic performance of corroded RC structures and the outcomes of this study are expected to provide a useful reference for the seismic risk management of coastal cities.

Massive short-term passenger flow prediction models of urban rail transit stations have been used in different conditions. However, researchers encountered several challenges while selecting the optimal passenger flow information input matrix and eliminating the redundant information in the original data. In this paper, we propose a learning network based on the optimal passenger flow input information algorithm (MTFLN) method. Based on the passenger flow information attribute of the predicted target station and the correlation coefficient distribution characteristics in different stages, the parameters of the optimal passenger flow information input algorithm (OPFIIA) are set reasonably. The optimal passenger flow input information matrix is also determined. We combine the training results of MTFLN method to optimize parameters of the Adam optimization algorithm-long and short-time memory network (Adam-LSTM). In the case study, the short-term passenger flow predicted values of urban rail transit stations are presented by three predicted methods including Adam-LSTM, Elman neural network and autoregressive integrated moving average (ARIMA). The presented results demonstrate that selecting the initial time correlation matrix by combining the attributes of the predicted target station and using OPFIIA method can not only effectively select the optimal passenger flow input information matrix, but also improve training efficiency and the prediction accuracy of traditional predicted model.

Blackleg (Phoma stem canker), caused by the Leptosphaeria species complex L. maculans and L. biglobosa, is a severe and economically important disease in most oilseed production areas worldwide. In this study, the virulence structure of a population of 115 L. maculans isolates from six counties in Alberta, Canada, was determined on three differential canola cultivars, Westar, Quinta and Glacier. Based on the differential interactions as assessed on a 0-9 severity scale, the isolates were classified into four pathogenicity groups: PG-2, PG-T, PG-3, and PG-4. When the differential interactions were assessed by analysis of variance of lesion size, the isolates were classified as into PG-2, PG-T and PG-4, among which PG-4 was predominant comprising 59% of the isolates, followed by PG-T at 37% and PG-2 at 4%. The presence or absence of five avirulence genes, AvrLm1, AvrLm4-7, AvrLm6, AvrLmJ1 and AvrLm11, in each of the isolates was evaluated by PCR analysis. AvrLm1 was absent, while AvrLm4-7 and AvrLm6 were present, in all of the 115 isolates. AvrLmJ1 and AvrLm11 were present in most isolates but absent in three isolates collected from Ponoka County. AvrLm11 also was absent in another two isolates collected from Strathcona County. This study describes the current status of virulence in L. maculans population in Alberta, which may provide guidance for blackleg resistance breeding efforts and cultivar deployment. [ABSTRACT FROM AUTHOR]