关键词： 非编码RNA-蛋白质相互作用   LightGBM   随机森林   极端梯度增强算法   卷积自编码器  

摘要： 目的非编码RNA-蛋白质的相互作用(noncoding RNA-protein interactions,ncRPI)具有重要的生物学意义,目前预测其相互作用已成为当下研究非编码RNA (noncoding RNA,ncRNA)和蛋白质功能的重要途径之一。方法本研究基于ncRNA和蛋白质的序列信息提取特征,运用卷积自编码器预处理原始数据,训练三个机器学习模型:LightGBM(LBM)、随机森林(random forest,RF)和极端梯度增强算法(extreme gradient boosting,XGB),预测ncRNA与蛋白质的相互作用。结果在RPI369和RPI488两个数据集做5倍交叉验证,LBM、RF与XGB三个模型在两个数据集均达到较高的预测准确率,在RPI369数据集三个模型的预测准确率分别为0.757(LBM)、0.791(RF)、0.791(XGB),在RPI488数据集三个模型的预测准确率分别为0.918(LBM)、0.908(RF)、0.918(XGB);三个模型在RPI1807、RPI2241、RPI13254大数据集也取得较高的AUC(area under curve)值,在RPI1807三个模型的AUC值均为0.99,在RPI2241三个模型最低AUC值为0.87,在RPI13254三个模型最低AUC值为0.81,都表现出较好的预测准确性。结论机器学习方法能够预测ncRNA与蛋白质是否存在相互作用。

关键词： Fused features   LBP   Pixel based results   PF clustering   Weiner Filter  

摘要： Background and Objective: Brain tumor occurs because of anomalous development of cells. It is one of the major reasons of death in adults around the globe. Millions of deaths can be prevented through early detection of brain tumor. Earlier brain tumor detection using Magnetic Resonance Imaging (MRI) may increase patient's survival rate. In MRI, tumor is shown more clearly that helps in the process of further treatment. This work aims to detect tumor at an early phase. Methods: In this manuscript, Weiner filter with different wavelet bands is used to de-noise and enhance the input slices. Subsets of tumor pixels are found with Potential Field (PF) clustering. Furthermore, global threshold and different mathematical morphology operations are used to isolate the tumor region in Fluid Attenuated Inversion Recovery (Flair) and T2 MRI. For accurate classification, Local Binary Pattern (LBP) and Gabor Wavelet Transform (GWT) features are fused. Results: The proposed approach is evaluated in terms of peak signal to noise ratio (PSNR), mean squared error (MSE) and structured similarity index (SSIM) yielding results as 76.38, 0.037 and 0.98 on T2 and 76.2, 0.039 and 0.98 on Flair respectively. The segmentation results have been evaluated based on pixels, individual features and fused features. At pixels level, the comparison of proposed approach is done with ground truth slices and also validated in terms of foreground (FG) pixels, background (BG) pixels, error region (ER) and pixel quality (Q). The approach achieved 0.93 FG and 0.98 BG precision and 0.010 ER on a local dataset. On multimodal brain tumor segmentation challenge dataset BRATS 2013, 0.93 FG and 0.99 BG precision and 0.005 ER are acquired. Similarly on BRATS 2015, 0.97 FG and 0.98 BG precision and 0.015 ER are obtained. In terms of quality, the average Q value and deviation are 0.88 and 0.017. At the fused feature based level, specificity, sensitivity, accuracy, area under the curve (AUC) and dice similarity coef

关键词： Outcome prediction   Parkinson's disease   Montreal cognitive assessment (MoCA)   Predictor algorithms   Feature selection  

摘要： Background: Given the increasing recognition of the significance of non-motor symptoms in Parkinson's disease, we investigate the optimal use of machine learning methods for the prediction of the Montreal Cognitive Assessment (MoCA) score at year 4 from longitudinal data obtained at years 0 and 1. Methods: We selected n = 184 PD subjects from the Parkinson's Progressive Marker Initiative (PPMI) database (93 features). A range of robust predictor algorithms (accompanied with automated machine learning hyperparameter tuning) and feature subset selector algorithms (FSSAs) were selected. We utilized 65%, 5% and 30% of patients in each arrangement for training, training validation and final testing respectively (10 randomized arrangements). For further testing, we enrolled 308 additional patients. Results: First, we employed 10 predictor algorithms, provided with all 93 features;an error of 1.83 +/- 0.13 was obtained by LASSOLAR (Least Absolute Shrinkage and Selection Operator - Least Angle Regression). Subsequently, we used feature subset selection followed by predictor algorithms. GA (Genetic Algorithm) selected 18 features;subsequently LOLIMOT (Local Linear Model Trees) reached an error of 1.70 +/- 0.10. DE (Differential evolution) also selected 18 features and coupled with Thiel-Sen regression arrived at a similar performance. NSGAII (Non-dominated sorting genetic algorithm) yielded the best performance: it selected six vital features, which combined with LOLIMOT reached an error of 1.68 +/- 0.12. Finally, using this last approach on independent test data, we reached an error of 1.65. Conclusion: By employing appropriate optimization tools (including automated hyperparameter tuning), it is possible to improve prediction of cognitive outcome. Overall, we conclude that optimal utilization of FSSAs and predictor algorithms can produce very good prediction of cognitive outcome in PD patients.

关键词： Technical state   circuit breaker   short circuit currents   technical diagnostics   machine learning   pattern recognition   gradient boosted trees algorithm   sample  

摘要： This paper is devoted to improving the assessment of the technical state (health index) of power circuit breakers through the use of machine learning, which allows to formalize expert knowledge, to identify implicit correlations and dependencies, and to automate the decision-making process under uncertainty. The paper proves the possibility of using machine learning based on decision trees to solve the task of assessing the power circuit breakers technical state and solves the problem of determining its optimal parameters and the training sample. The authors of the study carried out approbation of the developed model for assessing the technical state of breakers based on analyzing a large area of electrical networks in the Sverdlovsk Region, the Russian Federation, which proved its efficiency. The necessity of taking into account short-circuit currents when recognizing the pattern of the power breakers technical state is substantiated.

关键词： Air pollution   PM2.5 components   Geostatistical interpolation   Machine learning  

摘要： Numerous modeling approaches to estimate concentrations of PM2.5 components have been developed to derive better exposures for health studies, including geostatistical interpolation approaches, land use regression models and, models based on remote sensing technology. Recently, there have been some efforts to develop models based on machine learning algorithms. Each one of these exposure assessment methods has inherent uncertainties resulting in varying levels of exposure misclassification. To date, only a few studies have attempted to systematically compare exposure estimates from different PM2.5 constituent models. Our research addresses this gap, by comparing the predictive capabilities of ordinary geostatistical interpolation (Ordinary Kriging - OK), hybrid interpolation (combination of Empirical Bayesian Kriging and land use regression), and machine learning techniques (forest-based regression) for estimating PM2.5 constituents in Eastern Massachusetts in the United States. We compared the estimates of 10 ambient PM2.5 components, which included Al, Cu, Fe, K, Ni, Pb, S, Ti, V, and Zn. The OK model performed poorest for all PM2.5 components, with an R-2 under 0.30. The hybrid model presented a slight improvement, especially for Cu and Fe, for which the R-2 value increased to 0.62 and 0.59, respectively. These elements presented the highest R-2 value from the hybrid model. The forest model presented the best performance, with R-2 values higher than 0.7 for most of the particle components, including Cu, Fe, Ni, Pb, Ti, and V. Same as observed with the hybrid model, the forest model for Cu and Fe explained the highest concentration variance, with a R-2 value equal to 0.88 and 0.92, respectively. The forest model for K, S, and Zn performed poorest with an R-2 value of 0.54, 0.37, and 0.44, respectively. The results presented here can be useful for the environmental health community to more accurately estimate PM2.5 constituents over space.

关键词： Oxidoreductase subfamily classes   standardization normal distribution PSSM   correlation coefficient PSSM   RFE-SND-CC-PSSM   recursive feature elimination   SVM  

摘要： Oxidoreductase is an enzyme that widely exists in organisms. It plays an important role in cellular energy metabolism and biotransformation processes. Oxidoreductases have many subclasses with different functions, creating an important classification task in bioinformatics. In this paper, a dataset of 2640 oxidoreductase sequences was used to perform an analysis and comparison. The idea of dipeptides was introduced to process the Position Specific Score Matrix (PSSM), since each dipeptide consists of two amino acids and each column of PSSM corresponds to the information of one amino acid. Two kinds of dipeptide scores were proposed, the Standardization Normal Distribution PSSM (SND-PSSM) and the Correlation Coefficient PSSM (CC-PSSM). Recursive Feature Elimination (RFE) is used to extract features from the SND-PSSM and CC-PSSM, and the two sets of extracted features are combined to form a new feature matrix, the RFE-SND-CC-PSSM. The results show that, with the proposed method and a kernel-based nonlinear SVM classifier, the accuracy can reach 95.56% by the Jackknife test. Our method greatly improves the accuracy of oxidoreductase subclass prediction. Using this method to predict the categories of the 6 major types of enzymes effectively improves its prediction accuracy to 94.54%, indicating that this method has general applicability to other protein problems. The results show that our method is effective and universally applicable, and might be complementary to the existing methods.

关键词： Electrical vehicle   Heat pump   Refrigerant injection   Machine learning   SVR   Adaboost.R2  

摘要： A method to predict the performance of R134a heat pump with EVI (Economized Vapor Injection), is proposed in the present study. Models using SVR (Support Vector Regression) as the base estimator and Adaboost.R2 as the ensemble method, are established to predict the heating capacity and COP of the heat pump. Different feature sets for the model input are formed, based on the working principle of the heat pump system and correlation analysis. Parameters of the models are optimized to improve prediction performance. The simulation results are compared with the experimental results, and the relative errors for heating capacity and COP prediction are within 8.5%. Moreover, the impacts of injection pressure on the EVI heat pump system are discussed and simulated using the model established. The optimum injection pressure of the heat pump system can be obtained from the model under different working conditions.

关键词： Satellites   Telemetry   Space vehicles   Recurrent neural networks   Monitoring   Data mining   Machine learning  

摘要： Spacecrafts are critical systems that have to survive space environment effects. Due to its complexity, these types of systems are designed in away to mitigate errors and maneuver the critical situations. Spacecraft delivers to the ground operator an abundance data related to system status telemetry;the telemetry parameters are monitored to indicate spacecraft performance. Recently, researchers proposed using Machine Learning (ML)/Telemetry Mining (TM) techniques for telemetry parameters forecasting. Telemetry processing facilitates the data visualization to enable operators understanding the behavior of the satellite in order to reduce failure risks. In this paper, we introduce a comparison between the different machine learning techniques that can be applied for low earth orbit satellite telemetry mining. The techniques are evaluated on the bases of calculating the prediction accuracy using mean error and correlation estimation. We used telemetry data received from Egyptsat-1 satellite including parameters such as battery temperature, power bus voltage and load current. The research summarizes the performance of processing telemetry data using autoregressive integrated moving average (ARIMA), Multilayer Perceptron (MLP), Recurrent Neural Network (RNN), Long Short-Term Memory Recurrent Neural Network (LSTM RNN), Deep Long Short-Term Memory Recurrent Neural Networks (DLSTM RNNs), Gated Recurrent Unit Recurrent Neural Network (GRU RNN), and Deep Gated Recurrent Unit Recurrent Neural Networks (DGRU RNNs).

关键词： accretion, accretion discs   methods: data analysis   methods: statistical   astronomical data bases: miscellaneous   stars: evolution   stars: pre-main-sequence  

摘要： The second Gaia Data Release (DR2) contains astrometric and photometric data for more than 1.6 billion objects with mean Gaia G magnitude <20.7, including many Young Stellar Objects (YSOs) in different evolutionary stages. In order to explore the YSO population of the Milky Way, we combined the Gaia DR2 data base with Wide-field Infrared Survey Explorer (WISE) and Planck measurements and made an all-sky probabilistic catalogue of YSOs using machine learning techniques, such as Support Vector Machines, Random Forests, or Neural Networks. Our input catalogue contains 103 million objects from the DR2xAllWISE cross-match table. We classified each object into four main classes: YSOs, extragalactic objects, main-sequence stars, and evolved stars. At a 90 per cent probability threshold, we identified 1 129 295 YSO candidates. To demonstrate the quality and potential of our YSO catalogue, here we present two applications of it. (1) We explore the 3D structure of the Orion A star-forming complex and show that the spatial distribution of the YSOs classified by our procedure is in agreement with recent results from the literature. (2) We use our catalogue to classify published Gaia Science Alerts. As Gaia measures the sources at multiple epochs, it can efficiently discover transient events, including sudden brightness changes of YSOs caused by dynamic processes of their circumstellar disc. However, in many cases the physical nature of the published alert sources are not known. A cross-check with our new catalogue shows that about 30 per cent more of the published Gaia alerts can most likely be attributed to YSO activity. The catalogue can be also useful to identify YSOs among future Gaia alerts.

关键词： Depression   EEG   Ensemble model   Deep learning  

摘要： The diagnosis of depression almost exclusively depends on doctor-patient communication and scale analysis, which have the obvious disadvantages such as patient denial, poor sensitivity, subjective biases and inaccuracy. An objective, automated method that predicts clinical outcomes in depression is essential for increasing the accuracy of depression recognition and treatments. This paper aims at better recognizing depression using the transformation of EEG features and machine learning methods. An experiment based on emotional face stimuli task was conducted, and twenty-eight subjects' EEG data were recorded from 128-channel HydroCel Geodesic Sensor Net (HCGSN) by Net Station software. The Mini International Neuropsychiatric Interview (MINI) was used by psychiatrists as the criterion for diagnosis of depression patients. The power spectral density and activity were respectively extracted as original features using Auto-regress model and Hjorth algorithm with different time windows. Two separate approaches processed the features: ensemble learning and deep learning. For the ensemble learning, a deep forest transformed the original features to new features that potentially improve feature engineering and a support vector machine (SVM) that was applied as classifier. For deep learning method, we added spatial information of EEG caps to both features by image conversion and adopted convolutional neural network (CNN) to recognize them. The performance of both methods was evaluated for separated and total frequency bands. As a result, the best accuracy obtained was 89.02% when we used the ensemble model and power spectral density. The best accuracy of deep learning method was 84.75% using the activity. These experimental results prove the efficiency of the proposed methods and show that EEG could be used as a reliable indicator for depression recognition, which makes it possible for EEG-based portable system design and application in auxiliary depression recognition in th