Prediction Accuracy Of Method Research Articles

Adaptive Multi-Kernel Graph Neural Network for Drug-Drug Interaction Prediction.

Combination therapy, which synergistically enhances treatment efficacy and inhibits disease progression through the combined effects of multiple drugs, has emerged as a mainstream approach for treating complex diseases and alleviating symptoms. However, drug-drug interactions (DDIs) can sometimes lead to adverse reactions, potentially endangering lives. Therefore, developing efficient and accurate DDI prediction methods is crucial for elucidating drug mechanisms and preventing side effects. Current prediction methods often focus solely on the presence of interactions between drugs when constructing DDI graphs, neglecting the specific types of DDIs. This oversight can result in a decline in predictive performance. To address this issue, we propose an Adaptive Multi-Kernel Graph Neural Network (AMKGNN) for DDI prediction. AMKGNN differentiates DDIs into increase-type and decrease-type interactions, constructing separate increased DDI and decreased DDI graphs as convolutional kernels. AMKGNN employs a graph kernel learning mechanism that adaptively determines the optimal threshold between high-frequency and low-frequency signals in the network to capture node embeddings. Initially, AMKGNN learns drug embedding representations based on these two graph convolutional kernels and various drug features. These representations are then concatenated and input into a deep neural network to predict potential DDIs. The results show that our model achieved AUC and AUPR values above 90% across three sub-tasks on two datasets, significantly outperforming the other five comparison models. Furthermore, ablation experiments and case studies validate the superiority of AMKGNN.

Machine Learning and Deep Learning Techniques for EEG-Based Prediction of Psychiatric Disorders

Early detection of psychiatric disorders as well as efficient treatment are difficult owing to their challenges, which require accurate prediction methods in healthcare. When combined with ML and DL techniques, EEG data promises to yield a promising method for enhancing diagnostic accuracy. In this study, the performance of a wide spectrum of ML and DL techniques for predicting psychiatric disorders from EEG datasets is evaluated and the best choice is found for a particular condition. The study carried an analysis based on public datasets representing diverse psychiatric disorders through systematic analysis. Advanced DL architectures comprising of CNNs and RNNs were compared against the classical traditional ML techniques such as RlForest and Support Vector Machines (SVMs). A comparison between these models was made based on key performance metrics such as accuracy, sensitivity, and specificity. Results showed that DL models, particularly CNNs, excel at feature extraction and classification over traditional ML methods with their highest accuracy of predicting major depressive disorder above 92%. But ML techniques were able to complete faster computationally, in spite of slightly lower predictive accuracy. As DL models excel at capturing complex patterns within EEG data, these findings suggest that there are increased computational demands associated with them. Following that, advanced pattern recognition capabilities associated with DL techniques benefit substantially from the predictive modeling offered by EEG, although their computational efficiency presents as a limitation. This study highlights the importance of hybrid methods combining the best properties of both ML and DL for psychiatric disorders prediction to get improved accuracy and scalability, which is conditioning this generation of safer diagnostic tools for clinical practice .

DisDock: A Deep Learning Method for Metal Ion-Protein Redocking.

The structures of metalloproteins are essential for comprehending their functions and interactions. The breakthrough of AlphaFold has made it possible to predict protein structures with experimental accuracy. However, the type of metal ion that a metalloprotein binds and the binding structure are still not readily available, even with the predicted protein structure. In this study, we present DisDock, a deep learning method for predicting protein-metal docking. DisDock takes distogram of randomly initialized protein-ligand configuration as input and outputs the distogram of the predicted binding complex. It combines the U-net architecture with self-attention modules to enhance model performance. Taking inspiration from the physical principle that atoms in closer proximity display a stronger mutual attraction, this predictor capitalizes on geometric information to uncover latent characteristics indicative of atom interactions. To train our model, we employ a high-quality metalloprotein dataset sourced from the Mother of All Databases (MOAD). Experimental results demonstrate that our approach outperforms other existing methods in prediction accuracy for various types of metal ions.

Deep learning-based differential gut flora for prediction of Parkinson's.

There had been extensive research on the role of the gut microbiota in human health and disease. Increasing evidence suggested that the gut-brain axis played a crucial role in Parkinson's disease, with changes in the gut microbiota speculated to be involved in the pathogenesis of Parkinson's disease or interfere with its treatment. However, studies utilizing deep learning methods to predict Parkinson's disease through the gut microbiota were still limited. Therefore, the goal of this study was to develop an efficient and accurate prediction method based on deep learning by thoroughly analyzing gut microbiota data to achieve the diagnosis of Parkinson's disease. This study proposed a method for predicting Parkinson's disease using differential gut microbiota, named the Parkinson Gut Prediction Method (PGPM). Initially, differential gut microbiota data were extracted from 39 Parkinson's disease (PD) patients and their corresponding 39 healthy spouses. Subsequently, a preprocessing method called CRFS (combined ranking using random forest scores and principal component analysis contributions) was introduced for feature selection. Following this, the proposed LSIM (LSTM-penultimate to SVM Input Method) approach was utilized for classifying Parkinson's patients. Finally, a soft voting mechanism was employed to predict Parkinson's disease patients. The research results demonstrated that the Parkinson gut prediction method (PGPM), which utilized differential gut microbiota, performed excellently. The method achieved a mean accuracy (ACC) of 0.85, an area under the curve (AUC) of 0.92, and a receiver operating characteristic (ROC) score of 0.92. In summary, this method demonstrated excellent performance in predicting Parkinson's disease, allowing for more accurate predictions of Parkinson's disease.

Applicability of AlphaFold2 in the modeling of dimeric, trimeric, and tetrameric coiled-coil domains.

Coiled coils are a common protein structural motif involved in cellular functions ranging from mediating protein-protein interactions to facilitating processes such as signal transduction or regulation of gene expression. They are formed by two or more alpha helices that wind around a central axis to form a buried hydrophobic core. Various forms of coiled-coil bundles have been reported, each characterized by the number, orientation, and degree of winding of the constituent helices. This variability is underpinned by short sequence repeats that form coiled coils and whose properties determine both their overall topology and the local geometry of the hydrophobic core. The strikingly repetitive sequence has enabled the development of accurate sequence-based coiled-coil prediction methods; however, the modeling of coiled-coil domains remains a challenging task. In this work, we evaluated the accuracy of AlphaFold2 in modeling coiled-coil domains, both in modeling local geometry and in predicting global topological properties. Furthermore, we show that the prediction of the oligomeric state of coiled-coil bundles can be achieved by using the internal representations of AlphaFold2, with a performance better than any previous state-of-the-art method (code available at https://github.com/labstructbioinf/dc2_oligo).

Advanced predictive model and feature importance analysis for geological characteristics in tunnelling operations

Geological characteristics (GC) play a pivotal role in the efficiency and safety of tunnelling operations, necessitating accurate prediction methods. This study utilizes a dataset comprising operational parameters such as mean thrust (MT), penetration rate (PR), field penetration index (FPI), torque penetration index (TPI), advance rate (AR), mean cutterhead torque (MCT), and specific energy (SE). Bayesian optimization (BO) is employed for hyperparameter tuning of the extra trees (ET) algorithm to enhance prediction accuracy. The model was tasked with predicting three GCs using the operational parameters. The ET-BO model achieves near-perfect performance across all three GC classes, with an F1 score and accuracy of over 95%, during training. In testing, the model maintains strong performance with an F1 score and accuracy of over 91%, demonstrating robust predictive capability across all classes.

An AutoML-Powered Analysis Framework for Forest Fire Forecasting: Adapting to Climate Change Dynamics

Wildfires pose a serious threat to ecosystems and human safety, and with the backdrop of global climate change, the prediction of forest fires has become increasingly important. Traditional machine learning methods face challenges in forest fire prediction, such as difficulty identifying feature parameters, manual intervention in model selection, and hyperparameter tuning, which affect prediction accuracy and efficiency. This study proposes an analytical framework for forest fire prediction based on Automated Machine Learning (AutoML) technology to address the challenges traditional machine learning methods face in forest fire prediction. We collected meteorological, topographical, and vegetation data from Guangxi Province, with meteorological data covering 1994 to 2023, providing comprehensive background information for our prediction model. Using the prediction model, which was constructed with the AutoGluon framework, the experimental results indicate that models under the AutoGluon framework (e.g., KNeighborsDist classifier) significantly outperform traditional machine learning models in terms of accuracy, precision, recall, and F1-Score, with the highest accuracy rate reaching 0.960. Model error analysis shows that models under the AutoGluon framework perform better in error control. This study provides an efficient and accurate method for forest fire prediction, which is of great significance for decision-making in forest fire management and for protecting forest resources and ecological security.

A Preliminary Neural Network-Based Composite Method for Accurate Prediction of Enthalpies of Formation.

A composite method, named ANN-G3S, is introduced, adapting from G3S theory and employing distinct sets of multiplicative scale factors. An artificial neural network (ANN)-based classification model is utilized to select optimal sets of four scale factors for electronic correlation and basis set expansion terms in electronic systems. The correlation and basis set terms are scaled by four parameters, two for atoms and the other two for molecules. The ANN model is trained on the G3/05 test set to identify the best parameter set for each electronic system. To validate the method, 10% of the structures from the test set are randomly excluded from training and optimization, forming a separate validation set. The method demonstrates a mean deviation of 1.11 kcal mol-1 for the G3/05 set and 0.89 kcal mol-1 for the validation set, close to the value presented by the G4 method and surpassing the accuracy of the G3 method of 1.19 kcal mol-1 with significantly reduced computational cost. This method shows advantages by eliminating the need for purely empirical corrections, thereby enhancing both efficiency and accuracy in predicting heats of formation.

Prediction Method for Pomegranate Chlorophyll Content Based on Multi-feature Fusion of Unmanned Aerial Vehicle

The aim of this study was to obtain RGB and multispectral images of fruit tree canopy during the flowering period of pomegranate by multispectral unmanned aerial vehicle (UAV) to quickly and accurately predict the chlorophyll content in order to improve the monitoring efficiency of the orchard. A handheld chlorophyll meter was used to obtain the actual chlorophyll values, and image processing techniques were combined to extract parameters such as color features and texture features of the RGB images as well as vegetation index of the multispectral images. A chlorophyll content prediction method based on the support vector regression model and the convolutional neural network model (CNN-Attention) combined with the attention mechanism was established. The results showed that (1) the accuracy of the prediction model was improved by the fusion of RGB image features and multispectral image vegetation index. (2) After model comparison, the CNN-Attention model built under the fused features was the best in chlorophyll content prediction with R2, RE, and RMSE of 0.9699, 0.0052, and 0.6013, respectively.This study provides a more accurate method for fruit tree chlorophyll content prediction using UAVs, which provides a practical reference for orchard management.

Visualizing the Impact of Machine Learning on Cardiovascular Disease Prediction: A Comprehensive Analysis of Research Trends

Cardiovascular diseases (CVDs) continue to have a negative impact on global health, which highlights the need for accurate and efficient prediction methods. Machine learning (ML) techniques as tools for forecasting CVD has recently showed potential. This paper presents a comprehensive analysis of research trends in the field, focusing on visualizing the impact of ML in cardiovascular disease prediction. We used data visualization techniques to identify patterns and trends in an extensive database of scholarly publications on this subject that were published in Scopus between 1991 and 2023. The analysis reveals a substantial growth in research output, demonstrating the growing demand for ML-based CVD prediction. It reveals essential stakeholders and potential collaborators while highlighting the institutions and authors who have contributed most to this domain. The study also identifies high-impact journals that have published significant research in this domain, facilitating researchers in selecting appropriate outlets for dissemination. The study helps researchers identify the most critical areas for further research and fosters cooperation among subject-matter experts by offering insightful information about machine learning-based cardiovascular disease prediction development. The data is analyzed using the tools VOSviewer and Biblioshiny.

An Improved Decline Curve Analysis Method via Ensemble Learning for Shale Gas Reservoirs

As a clean unconventional energy source, shale gas reservoirs are increasingly important globally. Accurate prediction methods for shale gas production capacity can bring significant economic benefits by reducing construction and operating costs. Decline curve analysis (DCA) is an efficient method that uses mathematical formulas to describe production trends with minimal reliance on geological or engineering parameters. However, traditional DCA models often fail to capture the complex production dynamics of shale gas wells, especially in complex environments. To overcome these limitations, this study proposes an Improved DCA method that integrates multiple base empirical DCA models through ensemble learning. By combining the strengths of individual models, it offers a more robust and accurate prediction framework. We evaluated this method using data from 22 shale gas wells in region L, China, comparing it to six traditional DCA models, including Arps and the Logistic Growth Model (LGM). The results show that the Improved DCA model achieved superior performance—with an mean absolute error (MAE) of 0.0660, an mean squared error (MSE) of 0.0272, and an R2 value of 0.9882—and exhibited greater stability across various samples and conditions. This method provides a reliable tool for long-term production forecasting and optimization without extensive geological or engineering information.

Motion performance prediction of underwater gliders based on deep learning and image modeling

With the industrialization development and pedigree application, the underwater gliders are facing more extensive motion modes and application scenarios. However, it is complicated to establish a specific relationship between the explicit motion property and the implicit design parameters. This study proposes an accurate and rapid performance prediction method for underwater gliders based on deep learning and image modeling. First of all, a parametric model is established to generate large quantities of data about design parameters and performance indexes. Then, an image modeling method is employed to transform the numerical data of the glider body into structure images. Finally, a hybrid convolutional neural network-multilayer perceptron (CNN-MLP) model is proposed, and the structure images and other numerical data are input into the model to predict the total gliding range of the gliders. The results show that the developed hybrid CNN-MLP model achieves excellent prediction ability. Sea trial data further demonstrate the effectiveness of the proposed model. Thus, the proposed architecture provides a promising method for predicting underwater glider performance with multidimensional data and can be further applied to other types of gliders. Our research also provides a valuable reference for intelligent design and optimization iteration of other underwater equipment.

Vessel Trajectory Prediction Based on AIS Data: Dual-Path Spatial–Temporal Attention Network with Multi-Attribute Information

With the rapid growth of the global shipping industry, the increasing number of vessels has brought significant challenges to navigation safety and management. Vessel trajectory prediction technology plays a crucial role in route optimization and collision avoidance. However, current prediction methods face limitations when dealing with complex vessel interactions and multi-dimensional attribute information. Most models rely solely on global modeling in the temporal dimension, considering spatial interactions only later, failing to capture dynamic changes in trajectory interactions at different time points. Additionally, these methods do not fully utilize the multi-attribute information in AIS data, and the simple concatenation of attributes limits the model’s potential. To address these issues, this paper proposes a dual spacial–temporal attention network with multi-attribute information (DualSTMA). This network models vessel behavior and interactions through two distinct paths, comprehensively considering individual vessel intentions and dynamic interactions. Moreover, we divide vessel attributes into dynamic and static categories, with dynamic attributes fused during feature preprocessing, and with static attributes being controlled through a gating mechanism during spatial interactions to regulate the importance of neighboring vessel features. Benchmark tests on real AIS data show that DualSTMA significantly outperforms existing methods in prediction accuracy. Ablation studies and visual analyses further validate the model’s reliability and advantages.

Benchmark Investigation of SARS-CoV-2 Mutants' Immune Escape with 2B04 Murine Antibody: A Step Towards Unraveling a Larger Picture.

Even though COVID-19 is no longer the primary focus of the global scientific community, its high mutation rate (nearly 30 substitutions per year) poses a threat of a potential comeback. Effective vaccines have been developed and administered to the population, ending the pandemic. Nonetheless, reinfection by newly emerging subvariants, particularly the latest JN.1 strain, remains common. The rapid mutation of this virus demands a fast response from the scientific community in case of an emergency. While the immune escape of earlier variants was extensively investigated, one still needs a comprehensive understanding of how specific mutations, especially in the newest subvariants, influence the antigenic escape of the pathogen. Here, we tested comprehensive in silico approaches to identify methods for fast and accurate prediction of antibody neutralization by various mutants. As a benchmark, we modeled the complexes of the murine antibody 2B04, which neutralizes infection by preventing the SARS-CoV-2 spike glycoprotein's association with angiotensin-converting enzyme (ACE2). Complexes with the wild-type, B.1.1.7 Alpha, and B.1.427/429 Epsilon SARS-CoV-2 variants were used as positive controls, while complexes with the B.1.351 Beta, P.1 Gamma, B.1.617.2 Delta, B.1.617.1 Kappa, BA.1 Omicron, and the newest JN.1 Omicron variants were used as decoys. Three essentially different algorithms were employed: forced placement based on a template, followed by two steps of extended molecular dynamics simulations; protein-protein docking utilizing PIPER (an FFT-based method extended for use with pairwise interaction potentials); and the AlphaFold 3.0 model for complex structure prediction. Homology modeling was used to assess the 3D structure of the newly emerged JN.1 Omicron subvariant, whose crystallographic structure is not yet available in the Protein Database. After a careful comparison of these three approaches, we were able to identify the pros and cons of each method. Protein-protein docking yielded two false-positive results, while manual placement reinforced by molecular dynamics produced one false positive and one false negative. In contrast, AlphaFold resulted in only one doubtful result and a higher overall accuracy-to-time ratio. The reasons for inaccuracies and potential pitfalls of various approaches are carefully explained. In addition to a comparative analysis of methods, some mechanisms of immune escape are elucidated herein. This provides a critical foundation for improving the predictive accuracy of vaccine efficacy against new viral subvariants, introducing accurate methodologies, and pinpointing potential challenges.

Deep generative design of RNA aptamers using structural predictions.

RNAs represent a class of programmable biomolecules capable of performing diverse biological functions. Recent studies have developed accurate RNA three-dimensional structure prediction methods, which may enable new RNAs to be designed in a structure-guided manner. Here, we develop a structure-to-sequence deep learning platform for the de novo generative design of RNA aptamers. We show that our approach can design RNA aptamers that are predicted to be structurally similar, yet sequence dissimilar, to known light-up aptamers that fluoresce in the presence of small molecules. We experimentally validate several generated RNA aptamers to have fluorescent activity, show that these aptamers can be optimized for activity in silico, and find that they exhibit a mechanism of fluorescence similar to that of known light-up aptamers. Our results demonstrate how structural predictions can guide the targeted and resource-efficient design of new RNA sequences.

SPLENDID incorporates continuous genetic ancestry in biobank-scale data to improve polygenic risk prediction across diverse populations.

Polygenic risk scores are widely used in disease risk stratification, but their accuracy varies across diverse populations. Recent methods large-scale leverage multi-ancestry data to improve accuracy in under-represented populations but require labelling individuals by ancestry for prediction. This poses challenges for practical use, as clinical practices are typically not based on ancestry. We propose SPLENDID, a novel penalized regression framework for diverse biobank-scale data. Our method utilizes ancestry principal component interactions to model genetic ancestry as a continuum within a single prediction model for all ancestries, eliminating the need for discrete labels. In extensive simulations and analyses of 9 traits from the All of Us Research Program (N=224,364) and UK Biobank (N=340,140), SPLENDID significantly outperformed existing methods in prediction accuracy and model sparsity. By directly incorporating continuous genetic ancestry in model training, SPLENDID stands as a valuable tool for robust risk prediction across diverse populations and fairer clinical implementation.

Traffic Flow Prediction: A Method Using Bagging-Based Ensemble Learning Model

For the development of the national economy, transportation is strategically significant. As the increasing ownership of automobiles, traffic jams are a common occurrence. Accurate prediction of traffic flow contributes to diverting traffic effectively and improving the quality of urban traffic, in turn improving the operation of the overall transportation system. The rapid development of artificial intelligence technologies, especially machine learning and deep learning, has provided effective methods for accurate prediction of traffic flow. Based on the above, in order to improve the accuracy of the prediction and to extend the application of machine learning and deep learning in the prediction of traffic flow, this study proposed a bagging-based ensemble learning model. Firstly, normalization method is used to preprocess the data. Subsequently, base prediction models including decision tree, random forest, logistic regression, convolution neural network, long short-term memory and multilayer perceptron are selected for training the prediction model, respectively. Finally, bagging-based ensemble learning method is used to integrate these base prediction models to further predict traffic flow. The results of comparison between the single base prediction models and the bagging-based ensemble learning model on the five evaluation indicators show that, for predicting the traffic flow, the bagging-based ensemble learning model outperforms the base prediction models. Meanwhile, this study explores the potential in the application of machine learning, deep learning, and especially bagging-based ensemble learning to predict traffic flow.

Thermo-hydraulic performance of concentric tube heat exchangers with turbulent flow: Predictive correlations and iterative methods for pumping power and heat transfer

This research addresses the problem of predicting the thermo-hydraulic performance of concentric tube heat exchangers (CTHE) under turbulent flow conditions, a critical aspect in energy-efficient industrial systems such as HVAC, power generation, and chemical processing. Existing studies often lack accurate predictive methods for balancing heat transfer performance with pumping power requirements. To tackle this issue, novel correlations and an iterative Newton–Raphson method were developed for predicting pumping power and heat transfer rates. Three-dimensional CFD simulations of a water-to-water counter-flow CTHE were conducted, with Reynolds numbers ranging from 4000 to 8000 for both the hot and cold fluids. The simulations employed the Reynolds-Averaged Navier–Stokes (RANS) equations with the k−ω SST turbulence model. The results demonstrated that increasing the Reynolds number enhances both heat transfer rates and pumping power, with the cold fluid requiring consistently higher pumping power. New correlations were developed to predict pumping power, capturing the impact of both entry and fully developed flow regions. These correlations showed an average error of less than 2.33% when compared with the CFD data. The iterative Newton–Raphson method for predicting heat transfer rates demonstrated high accuracy, with an average error of 0.66% for heat transfer rate, 0.03% for hot fluid outlet temperature, and 0.01% for cold fluid outlet temperature. Additionally, we identified optimal operating conditions for efficient cooling and heating based on the heat capacity ratio (Cr). The novelty of this work lies in the development of new, highly accurate predictive correlations and iterative methods for optimizing CTHE performance, going beyond existing literature by providing comprehensive insights into the relationship between pumping power, heat transfer efficiency, and flow conditions.

Fragmentomics of plasma mitochondrial and nuclear DNA inform prognosis in COVID-19 patients with critical symptoms

BackgroundThe mortality rate of COVID-19 patients with critical symptoms is reported to be 40.5%. Early identification of patients with poor progression in the critical cohort is essential to timely clinical intervention and reduction of mortality. Although older age, chronic diseases, have been recognized as risk factors for COVID-19 mortality, we still lack an accurate prediction method for every patient. This study aimed to delve into the cell-free DNA fragmentomics of critically ill patients, and develop new promising biomarkers for identifying the patients with high mortality risk.MethodsWe utilized whole genome sequencing on the plasma cell-free DNA (cfDNA) from 33 COVID-19 patients with critical symptoms, whose outcomes were classified as survival (n = 16) and death (n = 17). Mitochondrial DNA (mtDNA) abundance and fragmentomic properties of cfDNA, including size profiles, ends motif and promoter coverages were interrogated and compared between survival and death groups.ResultsSignificantly decreased abundance (~ 76% reduction) and dramatically shorter fragment size of cell-free mtDNA were observed in deceased patients. Likewise, the deceased patients exhibited distinct end-motif patterns of cfDNA with an enhanced preference for “CC” started motifs, which are related to the activity of nuclease DNASE1L3. Several dysregulated genes involved in the COVID-19 progression-related pathways were further inferred from promoter coverages. These informative cfDNA features enabled a high PPV of 83.3% in predicting deceased patients in the critical cohort.ConclusionThe dysregulated biological processes observed in COVID-19 patients with fatal outcomes may contribute to abnormal release and modifications of plasma cfDNA. Our findings provided the feasibility of plasma cfDNA as a promising biomarker in the prognosis prediction in critically ill COVID-19 patients in clinical practice.

Operation prediction of open sun drying based on mathematical-physical model, drying kinetics and machine learning

In order to determine the moisture ratio of dried material whether the storage requirements are met, it was crucial to find an accurate prediction and convenient method in the open sun drying process. Therefore, the mathematical-physical model, drying dynamics model and machine learning methods were employed and compared in this study. The machine learning methods were first applied to predict the moisture ratio change of sweet potato during open sun drying. A large number of sweet potatoes drying experiments were carried out under open sun drying for theoretical analysis. The results shown that the drying kinetic model of sweet potato was also different under different drying climate conditions. The heat and mass transfer model of sweet potato was established and validated with R2 0.8990 and RMSE 0.0826. Different optimal machine learning prediction methods have be selected based on statistical metrics. Finaly, the machine learning prediction method was considered to be superior to the mathematical-physical model and the drying kinetic model in predicting moisture ratio. The results of this study can be analogized to drying process control of other agricultural products in the future.