A machine learning-based approach to ERα bioactivity and drug ADMET prediction

흡수, 분포, 대사, 배설 특성 및 독성을 예측하기 위한 효과적인 툴을 개발하는 것은 신약개발의 초기단계에서 NCE(new chemical entity)에 대한 가장 중요한 업무 중의 하나이다. 최근에 이런 시도중의 하나로서 ADME/T(absorption, distribution, metabolism, excretion, toxicity)관련 성질들의 예측에 support vector machine(SVM)을 이용하고 있다. 그리고 SVM은 ADME/T 성질들을 정확하게 예측하는데 많이 사용 되고 있다. 그러나 SVM 모델링에 두 가지 문제가 있다. 특성 선택(feature selection) 과 매개변수 설정(parameter setting)은 여전히 해결해야 할 과제이다. 이 두 가지 문제들은 SVM 분류의 효율성과 정확도에 결정적인 영향을 끼친다. 특히 특성 선택과 최적화된 SVM 변수의 설정은 서로 영향을 주기 때문에 동시에 다루어져야 한다. 여기서 우리는 genetic algorithm(GA) – 특성 선택에 사용 – 과 grid search(GS) method– 변수최적화에 사용 – 두 가지를 통합하는 효과적인 해결책을 제시하였다. ADME/T관련 성질 중 하나인 심장부정맥을 야기시키는 hERG 이온채널 저해제 분류 모델이 여기서 제안된 GA-GS-SVM을 위해 할당되고 테스트 되었다. 1891개의 화합물을 가지는 트레이닝 셋으로 단일 모델 3개, 앙상블 모델 3개, 총 6개의 모델을 만들었고 175개의 외부 데이터를 테스트 셋으로 사용하여 검증하였다. 데이터의 불균형 문제를 해결하기 위하여 GA-GS-SVM 단일 모델에 의한 예측 정확도와 GA-GS-SVM 앙상블 모델 예측 정확도를 비교하였으며, 앙상블모델을 사용하여 예측의 정확도를 높일 수 있었다. Developing effective tools for predicting absorption, distribution, metabolism, excretion properties and toxicity (ADME/T) of new chemical entities in the early stage of drug design is one of the most important tasks in drug discovery and development today. As one of these attempts, support vector machines (SVM) has recently been exploited for the prediction of ADME/T related properties. However, two problems in SVM modeling, i.e. feature selection and parameters setting, are still far from solved. The two problems have been shown to be crucial to the efficiency and accuracy of SVM classification. In particular, the feature selection and optimal SVM parameters setting influence each other, which indicates that they should be dealt with simultaneously. In this account, we present an integrated practical solution, in which genetic-based algorithm (GA) is used for feature selection and grid search (GS) method for parameters optimization. hERG ion-channel inhibitor classification models of ADME/T related properties has been built for assessing and testing the proposed GA-GS-SVM. We generated 6 different models that are 3 different single models and 3 different ensemble models using training set - 1891 compounds and validated with external test set - 175 compounds. We compared single model with ensemble model to solve data imbalance problems. It was able to improve accuracy of prediction to use ensemble model.

The ionospheric equivalent slab thickness is a key parameter for understanding the plasma distribution in the ionosphere, with direct relevance to satellite navigation, communication, and skywave over‐the‐horizon radar. However, traditional prediction methods often suffer from regional biases, limiting their global applicability. To address this, we develop a novel global prediction model for ionospheric slab thickness by integrating the Extreme Gradient Boosting (XGBoost) algorithm with Ensemble Learning (EL) techniques (XGBoost + EL). This hybrid approach leverages the strong nonlinear modeling capability of XGBoost and the robustness of EL to enhance prediction accuracy and generalization. The model was trained and tested on global occultation data from 2007 to 2020, incorporating key parameters such as vertical total electron content (vTEC), the F2‐layer critical frequency, and solar activity indices. Evaluation results demonstrate the model's high predictive accuracy, achieving a root mean square error (RMSE) of 62.3 km, a mean absolute error (MAE) of 41.5 km, a mean absolute percentage error (MAPE) of 13.1%, and a correlation coefficient ( R ) of 0.904. Notably, our model significantly outperforms the Neustrelitz equivalent Slab Thickness Model (NSTM), reducing RMSE, MAE, and MAPE by 54.8 km, 47.2 km, and 11.8%, respectively. It also shows superior performance over the International Reference Ionosphere (IRI) model, with corresponding reductions of 72.1 km, 57.1 km, and 14.7%. These findings underscore the considerable potential of machine learning for advancing global ionospheric modeling. Future work will focus on model optimization and assessing its applicability across diverse regions and temporal scales to further improve reliability.

Mapping of groundwater salinization and modelling using meta-heuristic algorithms for the coastal aquifer of eastern Saudi Arabia

Experiments have proven that traditional prediction research methods have limitations in practice. Proposing countermeasures for environmental changes is the key to optimal control of the cold chain environment and reducing the lag of control effects. In this paper, a cold chain transportation environment prediction method, combining k-means++ and a long short-term memory (LSTM) neural network, is proposed according to the characteristics of the cold chain transportation environment of agricultural products. The proposed prediction model can predict the trend of cold chain environment changes in the next ten minutes, which allows cold chain vehicle managers to issue control instructions to the environmental control equipment in advance. The fusion process for temperature and humidity data measured by multiple data sensors is performed with the k-means++ algorithm, and then the fused data are fed into an LSTM neural network for prediction based on time series. The prediction error of the prediction model proposed in this paper is very satisfactory, with a root-mean-square error (RMSE), mean absolute error (MAE), mean squared error (MSE), mean absolute percentage error (MAPE) and R-squared of 0.5707, 0.2484, 0.3258, 0.0312 and 0.9660, respectively, for temperature prediction, and with an RMSE, MAE, MSE, mean absolute percentage error and R-squared of 1.6015, 1.1770, 2.5648, 0.2736 and 0.9702, respectively, for humidity prediction. Finally, the LSTM neural network and back propagation (BP) neural network are compared in order to enhance the reliability of the results. In terms of the prediction effect of the temperature and humidity in cold chain vehicles transporting agricultural products, the proposed model has a higher prediction accuracy than that of existing models and can provide strategic support for the fine management and regulation of the cold chain transportation environment.

The current prediction model for the system efficiency of pumping units primarily relies on a mechanistic approach. However, this approach incorporates numerous unnecessary factors, thereby, increasing the cost associated with predictions. With the improvement of the oil field database, the available information is increasing. Some scholars propose a prediction model based on a single neural network, however, such models face challenges in effectively capturing complex data, resulting in lower prediction accuracy and limited resistance to interference. To solve the above problems, the study proposes a novel stacking integrated learning prediction model, which incorporates fivefold cross‐validation. First, the magnitude of the correlation coefficient was quantified using the Pearson correlation coefficient. Second, the impact and predictive features were normalized. Final, convolutional neural network (CNN), recurrent neural network (RNN), Long Short‐Term Memory network (LSTM), gated recurrent unit (GRU), and transformer are used as the base models, and fully connected neural network (FNN) is used as the metamodel. Each base model was trained by fivefold cross‐validation, and the predicted values of each fold were stacked by rows. Next, the predicted values of each base model are stacked by columns as input variables to the metamodel and metamodel learning is performed, and the stacking integrated learning prediction model based on fivefold crossover validation is established. To validate the accuracy of the model, we selected 5,000 actual well parameters, including 26 impact features and one predictive feature, for comparative analysis. This analysis presents the maximum percentage reduction in mean square error (MSE), mean absolute error (MAE), and root‐mean‐square error (RMSE) of our proposed integrated learning model concerning a single neural network prediction model as 28.26%, 24.40%, and 15.66%, respectively. The maximum percentage improvement in R2 is 17.74%. It shows that our proposed integrated learning prediction model has high prediction accuracy.

BackgroundCommonly used cardiovascular risk calculators do not provide risk estimation of stroke, a major postoperative complication with high morbidity and mortality. We developed and validated an accurate cardiovascular risk prediction tool for stroke, major cardiac complications (myocardial infarction or cardiac arrest), and mortality after non‐cardiac surgery.Methods and ResultsThis retrospective cohort study included 1 165 750 surgical patients over a 4‐year period (2007–2010) from the American College of Surgeons National Surgical Quality Improvement Program Database. A predictive model was developed with the following preoperative conditions: age, history of coronary artery disease, history of stroke, emergency surgery, preoperative serum sodium (≤130 mEq/L, >146 mEq/L), creatinine >1.8 mg/dL, hematocrit ≤27%, American Society of Anesthesiologists physical status class, and type of surgery. The model was trained using American College of Surgeons National Surgical Quality Improvement Program data from 2007 to 2009 (n=809 880) and tested using data from 2010 (n=355 870). Risk models were developed using multivariate logistic regression. The outcomes were postoperative 30‐day stroke, major cardiovascular events (myocardial infarction, cardiac arrest, or stroke), and 30‐day mortality. Major cardiac complications occurred in 0.66% (n=5332) of patients (myocardial infarction, 0.28%; cardiac arrest, 0.41%), postoperative stroke in 0.25% (n=2005); 30‐day mortality was 1.66% (n=13 484). The risk prediction model had high predictive accuracy with area under the receiver operating characteristic curve for stroke (training cohort=0.869, validation cohort=0.876), major cardiovascular events (training cohort=0.871, validation cohort=0.868), and 30‐day mortality (training cohort=0.922, validation cohort=0.925). Surgery types, history of stroke, and coronary artery disease are significant risk factors for stroke and major cardiac complications.ConclusionsPostoperative stroke, major cardiac complications, and 30‐day mortality can be predicted with high accuracy using this web‐based predictive model.

The aim of this study is to develop machine learning based models for staff churn and staff lifetime prediction. Three different approaches are used in the development of these models. In the first approach, prediction models are developed without feature selection. In the second approach, prediction models are developed using the Minimum Redundancy Maximum Relevance (mRMR) feature selection algorithm. In the third approach, prediction models are developed using the Principal Component Analysis (PCA) feature selection algorithm. Two different datasets are used in the development of the models. For predicting staff churn in both datasets, Logistic Regression (LR), Categorical Boosting (CatBoost), and Extreme Learning Machine (ELM) are used. To predict staff lifetime, Support Vector Machine (SVM), Gradient Boosting Machine (LightGBM), and K-Nearest Neighbors (KNN) are used. In order to evaluate the performance of the prediction models, Accuracy and F-Score are used for classification-based models, while Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) are used for regression-based models. The results obtained in this study show that the feature selection algorithms have no significant effect on the performance of the models.

This study presents a deep learning-based approach for forecasting product demand and predicting shipping delays within large-scale supply chain operations. Using a real-world dataset sourced from Kaggle comprising over 180,000 transaction-level records and 53 variables, two artificial neural network (ANN) models were developed: one for sales per customer (demand forecasting) and another for shipping delay estimation. After feature selection and preprocessing, eight key numerical variables including Product Price, Benefit per Order and Late Delivery Risk were standardized and used to train the models. The ANN architecture consisted of two hidden layers (64 and 32 neurons, respectively), with ReLU activation and dropout regularization. Experimental results show that the demand forecasting model achieved excellent performance, with a mean absolute error (MAE) of 11.72, root mean squared error (RMSE) of 23.94 and an R2 score of 0.97, indicating high predictive accuracy. In contrast, the shipping delay prediction model exhibited moderate results, with an MAE of 0.81, RMSE of 1.02 and R2 of 0.72, reflecting the underlying complexity of logistics dynamics. Visual analyses, including scatter plots, residual histograms and correlation heatmaps, confirmed strong linear relationships (e.g., Sales per Customer and Product Price: r = 0.78) and highlighted challenges due to heteroscedasticity and outliers. The study demonstrates the viability of ANN models in operational forecasting while identifying areas for enhancement particularly in delay prediction, where inclusion of temporal and categorical features could yield improved results. These findings contribute to data-driven supply chain optimization using modern Al techniques.

Effective prediction of aircraft failure rate has important guiding significance for formulating reasonable maintenance plans, carrying out reliable maintenance activities, improving health management levels, and ensuring the safety of aircraft flight, etc. Firstly, combining the advantages of time series model in eliminating random accidental factors interference, grey model in dealing with poor information, and the characteristics of artificial neural network in dealing with nonlinear data, the failure rate of aircraft equipment is predicted by ARIMA model, grey Verhulst model, and BP neural network model, and secondly, based on the idea of variable weight, the method of sum of squares of errors is used to reciprocate. Shapley value method and IOWA operator method determine the weighting coefficient and establish three combined forecasting models for aircraft failure rate prediction, so as to improve the accuracy of the algorithm. Finally, taking the data of actual aircraft failure rate as the research object, the performance indexes of design prediction model are judged by Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Index of Agreement (IA), Theil Inequality Coefficient (TIC), Equal Coefficient (EC), Nash-Sutcliffe Efficiency coefficient (NSE), Pearson test, and violin diagram of forecast error distribution. The experimental results show that: The forecasting precision of the combination model is better than that of the single model, and the evaluation index of combination forecasting model based on IOWA operator is better than that of other combination forecasting models, thus improving the forecasting accuracy and reliability. Compared with other typical prediction models simultaneously, it is verified that the proposed combined prediction model has strong applicability, high accuracy, and good stability, which provides a practical and effective technical method for aircraft fault prediction and has good application value.

Currently, the requirements of service quality in the electric power data network are getting higher and higher, and traffic prediction is an important premise to promote service quality. In order to accurately predict the total traffic of communication channels, a Multi-Applications Comprehensive Traffic Prediction (MACTP) model is proposed in this paper. Differing from F-ARIMA and S-ARIMA models which are used to predict the traffic of single application, the proposed MACTP model is used to predict the traffic of multi-applications conveyed in the channels. Simulation results show that MACTP model has higher accuracy and efficiency than classical prediction models, and it is suitable for electrical power data network.

With the in-depth implementation of China’s “National Strategy for Building a Strong Transportation Network,” the scale of expressway construction has continued to expand. As a result, the number of high-fill and deep-cut subgrade projects under complex geological conditions has increased significantly, leading to a surge in landslide-related issues. Consequently, accurate prediction of slope displacement is of critical importance for early warning and prevention of landslide disasters. This study proposes a hybrid prediction model, VMD-MPA-LSSVM-LSTM (VMLL), which integrates Variational Mode Decomposition (VMD), Marine Predators Algorithm (MPA), Least Squares Support Vector Machine (LSSVM), and Long Short-Term Memory (LSTM) networks. Using monitoring data from the high-fill embankment slope at Hongtuyao as the research subject, the VMLL model is employed to predict slope displacement based on small-sample data. The objective is to provide a more accurate method for early warning of slope displacement. Firstly, the original monitoring data are decomposed into trend displacement components and fluctuation displacement components using Variational Mode Decomposition (VMD). Subsequently, the trend component and the fluctuation component are predicted using Least Squares Support Vector Machine (LSSVM) and Long Short-Term Memory (LSTM) networks, respectively. Finally, the Marine Predators Algorithm (MPA) is employed to optimize the hyperparameters of the predictive models. Based on this framework, a VMLL-based slope displacement prediction model is constructed. To verify the superiority of the VMLL model, a comparative analysis was conducted against LSSVM, LSTM, and the VMD-LSSVM-LSTM models. The results demonstrate that the VMLL model achieves the highest prediction accuracy, with a Mean Absolute Percentage Error (MAPE) of 0.4022%, a Mean Absolute Error (MAE) of 0.016 mm, a Root Mean Square Error (RMSE) of 0.0206 mm, a coefficient of determination (R²) of 94.08%, and a Variance Accounted For (VAF) of 96.5%. Compared with the other three models, the VMLL model reduces the MAPE, MAE, and RMSE by 30.41–67.62%, 30.38–67.40%, and 27.32–71.00%, respectively. Meanwhile, it improves the R² and VAF by 5.95–217.51% and 0.38–11.48%, respectively. These improvements clearly demonstrate that the VMLL model outperforms the other models, indicating its significant advantage over single prediction models. Furthermore, three additional datasets were used to evaluate the model’s performance. The average values of the performance evaluation metrics for these datasets were 0.6207% (MAPE), 0.0238 mm(MAE), 0.0273 mm(RMSE), 90.33%(R²), and 96.81%(VAF), respectively. These results demonstrate the high accuracy and strong robustness of the proposed model in predicting both horizontal displacement and vertical settlement of slopes, providing a reliable methodological framework for slope stability assessment and landslide disaster early warning.

Water level prediction is an essential factor for the safe operation of pumping stations. However, due to the complex nonlinear relationship between the water level of the front pool of the pumping station and the influencing factors, the prediction of the water level is still inaccurate and untimely. Backpropagation (BP) neural network, improved particle swarm optimization-BP neural network (PSO-BP), support vector machine regression (SVR), and improved PSO-SVR were used to construct 4-h and 8-h ahead prediction models for pumping station prestation water levels. Mean absolute error, mean squared error (MSE), root mean square error (RMSE), mean absolute percentage error (MAPE), and R correlation coefficient were used as prediction evaluation metrics. This method is applied in the Baliwan Pumping Station, the highest pumping station in the South-to-North Water Diversion Eastern Route Project (SNWDERP). The results showed that the MSE, RMSE, and MAPE of the improved PSO-BP model were smaller than other models, whereas the R correlation coefficient was larger, confirming its high prediction accuracy. All models had higher prediction accuracy 4 h ahead than 8 h ahead. Combining the time-phased water level prediction method and hybrid machine learning can enhance the water level prediction accuracy of a pumping station pre-station.

Background: Cholangiocarcinoma is a molecularly heterogenous cancer arising from the biliary epithelium. Individual omic techniques have identified clinically relevant, targetable subtypes but most tumors have no targetable alterations based on exome or transcriptomic evaluation. Integrated multiomics approaches incorporating transcriptomic, proteomic, and phosphoproteomic characterization provide deeper understanding and identify non-mutated, activated pathways that can be targeted. To optimize the potential clinical utility of these approaches, minimal classifier features must be defined. Methods: Approximately 60,000 RNA, protein, and phosphoprotein features were extracted from 35 cholangiocarcinomas treated at the primary US study site. 206 patients treated and profiled at an institution in China were also included. Using the Multiomics Factor Analysis clustering method, patients were sorted into 3 distinct molecular profiles. Important molecular features were selected using competing feature selection (FS) methods: Boruta, RreliefF, variable selection using random forests (RF), recursive factor elimination (RFE), median decrease accuracy in RF, and Concrete Autoencoders. Machine learning classification prediction models including Extreme Gradient Boosting, gradient-boosted trees, multinomial logit, and support vector machine (SVM) were tested on each FS. Each combination of FS and predictive model was tested with 10 replicates of 10-fold cross-validation of a randomly-selected balanced test set and an oversampled, balanced training subset. Accuracy, F1 score, AUC, precision, and recall were assessed for each FS and predictive model combination. A custom function of physical assay cost and classification accuracy was optimized to find the best assay. KEGG pathway overrepresentation analysis was used to analyze feature subsets. Results: A consistent set of 50 important RNA and protein features was identified, including previously reported prognostic factors, like SpryD4 protein expression, as well as previously unreported pathway surrogates, such as Slc2A2 or Pipox protein. Pathway overrepresentation analysis found insulin signaling and resistance and amino acid biosynthesis/metabolism pathways were overrepresented in the discriminative feature set. The gradient-boosted trees algorithm combined with the RFE FS method had the best performance, with a multiclass macro average (±sd) cross-validation accuracy 89% (±6%), F1 score 89% (±8%), AUC 0.98 (±0.02), precision 90% (±10%), recall 89% (±6%). Conclusions: Feature selection methods and predictive machine learning models work synchronously to develop accurate molecular subtype assays based on just 50 protein and RNA levels. This strategy has the potential to make molecular subtyping for cholangiocarcinoma and other molecularly complex cancers fast and widely available. Citation Format: Ellen Larson, Erik Jessen, Dong-Gi Mun, Jennifer Tomlinson, Amro Abdelrahman, Danielle Carlson, Hojjat Salehinejad, Rory Smoot. Feature selection and machine learning strategies optimize an affordable molecular assay for cholangiocarcinoma subtype [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1107.

To develop and validate a standardized prediction model aiming at 1-year axial length elongation and to guide the orthokeratology lens practice. This retrospective study was based on medical records of myopic children treated with orthokeratology. Individuals aged 8-15years (n=1261) were included and divided into the primary cohort (n=757) and validation cohort (n=504). Feature selection was primarily performed to sort out influential predictors by high-throughput extraction. Then, the prediction model was developed using multivariable linear regression analysis completed by backward stepwise selection. Finally, the validation of the prediction model was performed by evaluation metrics (mean-square error, root-mean-square error, mean absolute error and ). No significant difference was found between primary and validation cohort (all p>0.05). After the feature selection, the crude model was adjusted by demographic information in multivariable linear regression analysis, and five final predictors were identified (all p<0.01). The interaction effect of age with 1-month change of zone-3mm flat K was detected (p<0.01); hence, two final prediction models were developed based on two age subgroups. The validation proved an acceptable performance. An effective multivariable prediction model aiming at 1-year axial length elongation was developed and validated. It can potentially help clinicians to predict orthokeratology efficacy and make valid adjustments. The influential variables revealed in this model can also provide designers directions to optimize the design of lens to improve the efficacy of myopia control.

The matter of success in forecasting precipitation is of great significance to flood control and drought relief, and water resources planning and management. For the nonlinear problem in forecasting precipitation time series, a hybrid prediction model based on variational mode decomposition (VMD) coupled with extreme learning machine (ELM) is proposed to reduce the difficulty in modeling monthly precipitation forecasting and improve the prediction accuracy. The monthly precipitation data in the past 60 years from Yan’an City and Huashan Mountain, Shaanxi Province, are used as cases to test this new hybrid model. First, the nonstationary monthly precipitation time series are decomposed into several relatively stable intrinsic mode functions (IMFs) by using VMD. Then, an ELM prediction model is established for each IMF. Next, the predicted values of these components are accumulated to obtain the final prediction results. Finally, three predictive indicators are adopted to measure the prediction accuracy of the proposed hybrid model, back propagation (BP) neural network, Elman neural network (Elman), ELM, and EMD-ELM models: mean absolute error (MAE), root mean squared error (RMSE), and mean absolute percentage error (MAPE). The experimental simulation results show that the proposed hybrid model has higher prediction accuracy and can be used to predict the monthly precipitation time series.