Network Approach Research Articles

Data Checking of Asymmetric Catalysis Literature Using a Graph Neural Network Approach

The range of chemical databases available has dramatically increased in recent years, but the reliability and quality of their data are often negatively affected by human-error fidelity. The size of chemical databases can make manual data curation/checking of such sets time consuming; thus, automated tools to help this process are highly desirable. Herein, we propose the use of Graph Neural Networks (GNNs) to identifying potential stereochemical misassignments in the primary asymmetric catalysis literature. Our method relies on the use of an ensemble of GNN models to predict the expected stereoselectivity of exemplars for a particular asymmetric reaction. When the majority of these models do not correlate to the reported outcome, the point is labeled as a possible stereochemical misassignment. Such identified cases are few in number and more easily investigated for their cause. We demonstrate the use of this approach to spot potential literature stereochemical misassignments in the ketone products resulting from catalytic asymmetric 1,4-addition of organoboron nucleophiles to Michael acceptors in two different databases, each one using a different family of chiral ligands (bisphosphine and diene ligands). Our results demonstrate that this methodology is useful for curation of medium-sized databases, speeding this process significantly compared to complete manual curation/checking. In the datasets investigated, human expert checking was reduced to 2.2% and 3.5% of the total data exemplars.

Directed Electrostatics Strategy Integrated as a Graph Neural Network Approach for Accelerated Cluster Structure Prediction.

We present a directed electrostatics strategy integrated as a graph neural network (DESIGNN) approach for predicting stable nanocluster structures on their potential energy surfaces (PESs). The DESIGNN approach is a graph neural network (GNN)-based model for building structures of large atomic clusters with specific sizes and point-group symmetry. This model assists in the structure building of atomic metal clusters by predicting molecular electrostatic potential (MESP) topography minima on their structural evolution paths. The DESIGNN approach is benchmarked on the prototype Mgn clusters with n < 150. The predicted MESP topography minima of Mgn clusters, n < 70, fairly agrees with the whole-molecule MESP topography calculations. Moreover, the ground-state structures of Mgn (n = 4-32) clusters generated through the DESIGNN approach corroborate well with the global minimum structures reported in the literature. Furthermore, this approach could generate novel symmetric isomers of medium to large Mgn clusters in the size regime, n < 150, by constraining the point-group symmetry of the parent clusters. The parent growth potential (GP) of a cluster gives a measure of its parent cluster to accommodate more atoms and characterize the structures on the DESIGNN-guided path. The GP of a cluster can also be interpreted as a measure of the cooperative interaction relative to its parent cluster. Along the highest GP paths, the DESIGNN approach is further employed to generate stable Mgn nanoclusters with n = 228, 236, 257, 260. Therefore, the DESIGNN approach holds great promise in accelerating the structure search and prediction of large metal clusters guided through MESP topography.

Self-consistent electron density with shell structure using neural network-based Pauli potential.

The orbital-free density functional theory (OF-DFT) based method is a convenient tool to carry out electronic structure calculations scaling almost linearly with the number of electrons. However, the main impediment in the application of this method is the unavailability of the accurate form for the non-interacting kinetic energy functional in terms of electron density. The Pauli kinetic energy functional is the unknown part of the kinetic energy functional, and the corresponding Pauli potential appears in the governing Euler equation. In the present study, we present a feed-forward neural network (NN) approach to represent the Pauli potential of a group of atomic systems possessing spherically symmetric ground-state densities. This NN-based representation of Pauli potential combined with the Hohenberg-Kohn variational principle yields self-consistent radial densities that accurately exhibit the correct atomic shell structure. For this approach, the electron density in the form of a grid serves as the input to the NN model. In addition, we calculated the non-interacting kinetic energy by summing the Pauli kinetic energy, derived from the NN-based Pauli potential, and the von Weizsäcker kinetic energy. Our results demonstrate high accuracy for smaller atoms, while larger atoms exhibit greater deviations when compared with smaller atoms. The method presented in this paper provides an efficient way to calculate the Pauli potential and the Pauli kinetic energy without the need for functional derivatives. Our study represents a significant step forward in the application of machine learning techniques to OF-DFT, showcasing the potential of NNs in improving the accuracy and efficiency of quantum mechanical calculations in atomic systems.

Revisiting the Metabolism of Donepezil in Rats Using Non-Targeted Metabolomics and Molecular Networking

Background/Objectives: Although donepezil, a reversible acetylcholinesterase inhibitor, has been in use since 1996, its metabolic characteristics remain poorly characterized. Therefore, this study aims to investigate the in vivo metabolism of donepezil using liquid chromatography–high-resolution mass spectrometry (LC-HRMS) based on a molecular networking (MN) approach integrated with a non-targeted metabolomics approach. Methods: After the oral administration of donepezil (30 mg/kg) in rats, urine, feces, and liver samples were collected for LC-HRMS analysis. Chromatographic and spectrometric data were processed through MN and multivariate data analysis to identify the in vivo metabolites of donepezil. Results: A total of 50 metabolites were characterized, including 23 newly identified metabolites. Donepezil was biotransformed by O-demethylation, N-debenzylation, and hydroxylation, and these metabolites are further conjugated with glucuronic acid and sulfurous acid. N-Desbenzyldonepezil (M4), didesmethyldonepezil (M5), and N-desbenzyldonepezil (M4) were identified as the most abundant metabolites in urine, feces, and liver samples, respectively. Conclusions: The metabolic characteristics of donepezil in rats were comparable to those in humans, indicating that a rat is a reliable model for studying donepezil metabolism. This study indicates that a MN approach combined with a metabolomics approach is a reliable tool to identify unknown metabolites of drugs and drug candidates.

Accelerating Benders Decomposition for sustainable food closed-loop supply chain network under uncertainty: a case study

PurposeThe purpose of this study was to address waste management in the food supply chain (FSC) through the integration of inspection processes in production and distribution centers under uncertain conditions, aiming to enhance sustainability across environmental, economic and social dimensions. The study introduces a sustainable forward and reverse FSC network using a closed-loop supply chain network approach to prevent the transfer of spoiled products, ultimately providing competitive advantages to stakeholders.Design/methodology/approachA robust multi-objective mathematical programming model is proposed, incorporating inspection processes to manage perishable products effectively. The model is solved using the Augmented Epsilon Constraint technique implemented in GAMS software, providing Pareto-optimal solutions tailored to decision-makers’ preferences. Furthermore, the methodology is applied in a real-world case study and solved with the Benders Decomposition algorithm to validate its practicality and effectiveness.FindingsThe proposed methodology effectively minimizes waste and enhances sustainability in the FSC by optimizing decision-making processes under uncertainty. The illustrative examples and real case study demonstrate the efficiency of the model and solution approach, highlighting the significant role of inspection in improving all three dimensions of sustainability.Practical implicationsThe study offers valuable insights into and tools for food industry managers to make informed strategic and tactical decisions. By addressing waste management through advanced supply chain modeling, the research helps organizations reduce costs, improve sustainability and gain a competitive edge in the market.Originality/valueThis research is novel in its focus on integrating inspection into the FSC network and addressing uncertainty through robust mathematical modeling. It contributes to the existing literature by demonstrating the impact of inspection on sustainability in FSCs and providing practical solutions for real-world implementation.

SymbolNet: Neural symbolic regression with adaptive dynamic pruning for compression

Abstract Compact symbolic expressions have been shown to be more efficient than neural network models in terms of resource consumption and inference speed when implemented on custom hardware such as FPGAs, while maintaining comparable accuracy.&amp;#xD;These capabilities are highly valuable in environments with stringent computational resource constraints, such as high-energy physics experiments at the CERN Large Hadron Collider.&amp;#xD;However, finding compact expressions for high-dimensional datasets remains challenging due to the inherent limitations of genetic programming, the search algorithm of most symbolic regression methods.&amp;#xD;Contrary to genetic programming, the neural network approach to symbolic regression offers scalability to high-dimensional inputs and leverages gradient methods for faster equation searching.&amp;#xD;Common ways of constraining expression complexity often involve multistage pruning with fine-tuning, which can result in significant performance loss.&amp;#xD;In this work, we propose $\tt{SymbolNet}$, a neural network approach to symbolic regression specifically designed as a model compression technique, aimed at enabling low-latency inference for high-dimensional inputs on custom hardware such as FPGAs.&amp;#xD;This framework allows dynamic pruning of model weights, input features, and mathematical operators in a single training process, where both training loss and expression complexity are optimized simultaneously.&amp;#xD;We introduce a sparsity regularization term for each pruning type, which can adaptively adjust its strength, leading to convergence at a target sparsity ratio.&amp;#xD;Unlike most existing symbolic regression methods that struggle with datasets containing more than $\mathcal{O}(10)$ inputs, we demonstrate the effectiveness of our model on the LHC jet tagging task (16 inputs), MNIST (784 inputs), and SVHN (3072 inputs).

Two-layer homolog network approach for PFAS nontarget screening and retrospective data mining

The rapid increase of novel per- and polyfluoroalkyl substances (PFAS) raises concerns, while their identification remains challenging. Here, we develop a two-layer homolog network approach for PFAS nontarget screening using mass spectrometry. The first layer constructs networks between homologs, with evaluation showing that it filters 94% of false candidates. The second layer builds a network between classes to expedite the identification of PFAS. We detected 94 PFAS in twelve waterproof products and two related industrial sludges, including 36 novel PFAS not previously reported in any sample. A local dataset is constructed for retrospective analysis by re-analyzing our previous samples, revealing fifteen novel PFAS in samples collected in 2005. The retrieval of the public database MassIVE uncovers novel PFAS in samples from seven countries. Here, we reveal the historic and global presence of novel PFAS, providing guidance for the management and policy-making concerning persistent chemicals.

A Spatio-Temporal Deep Learning Model for Automatic Arctic Sea Ice Classification with Sentinel-1 SAR Imagery

Arctic sea ice has a significant effect on global climate change, ship navigation, Arctic ecosystems, and human activities. Therefore, it is essential to produce high-resolution sea ice maps that accurately represent the geographical distribution of various sea ice types. Based on deep learning technology, many automatic sea ice classification algorithms have been developed using synthetic aperture radar (SAR) imagery over the last decade. However, sea ice classification faces two vital challenges: (1) it is difficult to distinguish sea ice types with close developmental stages solely from SAR images and (2) an imbalanced sea ice dataset has a significantly negative effect on ice classification model performance. In this article, a spatio-temporal deep learning model—the Dynamic Multi-Layer Perceptron (MLP)—is utilized to classify 10 sea ice types automatically. It consists of a SAR image branch and a spatio-temporal branch, which extracts SAR image features and spatio-temporal distribution characteristics of sea ice, respectively. By projecting similar image features to different positions in the spatio-temporal feature space dynamically, the Dynamic MLP model effectively distinguishes between similar sea ice types. Furthermore, to reduce the impact of data imbalance on model performance, the dynamic curriculum learning (DCL) method is used to train the Dynamic MLP model. Experimental results demonstrate that our proposed method outperforms the long short-term memory (LSTM) network approach in distinguishing between sea ice types with similar developmental stages. Moreover, the DCL training method can also effectively improve model performance in identifying minority ice types.

Deep Learning for Identification and Characterization of Ca ii Absorption Lines: A Multitask Convolutional Neural Network Approach

Quasar absorption lines are a powerful tool for studying the Universe, enabling us to probe distant gas, dust, and galaxy formation and evolution. However, detecting these lines, particularly Ca ii absorption lines, is a time-consuming and laborious process. Existing deep learning methods are prone to false positives and still require extensive manual verification and parameter measurement. This work presents three multitask convolutional neural network models and identifies the ResNet-CBAM model, which incorporates residual learning and an attention mechanism as the most effective. The results show that the ResNet-CBAM model achieves an accuracy of 99.7% in detecting Ca ii absorbers and excels in predicting critical parameters such as equivalent width and full width at half-maximum, with average correlation coefficients of 0.98 and 0.85, respectively. Furthermore, its remarkable generalization ability significantly improves detection precision on unseen data, rising from 20.3% of the cutting-edge model to 92.6%. In addition, with our numerous optimizations, our method can directly search for nonnormalized data, still achieving an accuracy of 98.6%. This translates to a dramatic reduction in manual inspection workload, paving the way for efficient and automated Ca ii absorber identification. In real-world applications on the Sloan Digital Sky Survey DR7 and DR12, our model successfully rediscovered 321 known Ca ii absorbers while identifying potential candidates in an additional 381 spectra. The codes used in this paper are available on Zenodo at doi:10.5281/zenodo.13953656.

Untargeted Characterization and Biological Activity of Amazonian Aqueous Stem Bark Extracts by Liquid and Gas Chromatography-Mass Spectrometry.

Aqueous stem bark extracts of Aspidosperma rigidum Rusby, Couroupita guianensis Aubl., Monteverdia laevis (Reissek) Biral, and Protium sagotianum Marchand have been reported as traditional remedies in several countries of the Amazonian region. Despite previous research, further investigation to characterize secondary metabolites and the biological activity of extracts is needed to derive potential applications. Metabolic profiling was carried out using liquid and gas chromatography coupled with mass spectrometry (UHPLC-MS/MS and GC-MS). The chemical composition of the studied plants was further compared by principal component analysis (PCA). Additionally, chemical profiles were correlated with antimicrobial and toxicity activities, which suggested potential metabolites for future research. We identified 16 and 32 compounds by UHPLC-MS/MS and GC-MS analysis, respectively. Antimicrobial activity was detected in three stem bark extracts. C. guianensis showed inhibition of all tested microorganisms, including antibiotic-resistant strains. Molecular networking approaches, in silico tools, and Pearson's correlation showed that antifungal compounds could be a terpene glycoside (r = 0.918) and/or a phenolic (r = 0.882) metabolite class. This study highlights the use of the established procedure in exploring the metabolomes of these species, which could be a novel source of antimicrobial drug discovery. Coupling the observed biological potential with UHPLC-MS/MS data has also accelerated the tracing of their bioactive compounds. These findings update the state of the art regarding the chemical composition and biological activity of the plant extracts, defining potential new applications for the pharmaceutical applications.

Enhancing Physical Spatial Resolution of Synthetic Aperture Sonar Images Based on Convolutional Neural Network

The sonar image has limitations on the physical spatial resolution due to system configuration and underwater environment, which often leads to challenges for underwater targets detection. Here, the deep learning method is applied to enhance the physical spatial resolution of underwater sonar images. Specifically, the U-shaped end-to-end neural network which contains down-sampling and up-sampling parts is proposed to improve the physical spatial resolution limited by the array aperture. The single target and multiple cases are considered separately. In both cases, the normalized loss on the testing sets declines rapidly, and the predicted high-resolution images own great agreement with the ground truth eventually. Further improvements in resolution are focused on, that is, compressing the predicted high-resolution image to its physical spatial resolution limitation. The results show that the trained end-to-end neural network could map high resolution targets to the impulse responses at the same location and amplitude with an uncertain target number. The proposed convolutional neural network approach could give a practical alternative to improve the physical spatial resolution of underwater sonar images.

Enhanced Prediction of CO2-Brine Interfacial Tension at Varying Temperature Using a Multibranch-Structure-Based Neural Network Approach.

Interfacial tension (IFTC-B) between CO2 and brine depends on chemical components in multiphase systems, intricately evolving with a change in temperature. In this study, we developed a convolutional neural network with a multibranch structure (MBCNN), which, in combination with a compiled data set containing measurement data of 1716 samples from 13 available literature sources at wide temperature and pressure ranges (273.15-473.15 K and 0-70 MPa), was used to quantitatively explore the correlation of various chemical components with IFTC-B at varying temperature, aiming to achieve accurate predictions of IFTC-B under complex conditions. Our multibranch neural network analysis yielded some important insights: (1) Leveraging the convolutional and multibranch structure, MBCNN effectively mitigates the adverse effects of sparse matrices resulting from the absence of certain basic components, exhibiting higher prediction accuracy particularly for low IFTC-B scenarios (MAE = 0.47, and R2 = 0.9921) than other AI models. (2) The multibranch structure allows MBCNN to additionally capture the interattribute relationship between temperature and each chemical component. Such interattribute relationships are quantitatively correlated with IFTC-B, demonstrating that varying temperature significantly influences the dependence of IFTC-B on chemical components in gas and brine by causing the variation in their solubility. Specifically, the ratio of IFTC-B to the molality of monovalent cations (Na+ and K+) and bivalent cations (Ca2+ and Mg2+) in brine, as well as to the mole fraction of non-CO2 components (CH4 and N2) in the gas phase, varies with increasing temperature, approximately following a quadratic function. (3) By formulating the effect of each attribute on IFTC-B and quantifying their respective weight, we derived a new piecewise function for predicting IFTC-B at three temperature intervals (T ≤ 293.15 K, 293.15 K < T ≤ 324.4 K, and T > 324.4 K), with high prediction performance (MAE = 2.3672, R2 = 0.9263) across a wide temperature range in saline aquifers.

Investigating continuous intention to use metaverse in higher education institutions: a dual-staged structural equation modeling-artificial neural network approach

The current study explores metaverse adoption among higher education institutions (HEIs) in the light of a theoretical framework to empower future perspectives of the metaverse as a learning platform. Even though this technology was just recently introduced to the higher education sector, very few attempts have been made to evaluate its impact. The purpose of this research is to analyze the elements that influence the continuous intention (CI) to utilize the metaverse technology in learning. The technology acceptance model (TAM) and the self-determination theory (SDT) are both included in this study. A questionnaire was developed and distributed to students attending private universities in order to obtain the data that was needed for the proposed model. Using a hybrid approach that consists of partial least squares structural equation modeling (PLS-SEM) and an artificial neural network (ANN) model, which combines a linear PLS model with compensation and a nonlinear ANN model without compensation, the effect of CI on using the metaverse as a learning platform is investigated. This approach was chosen because it contains both of these types of models. When it comes to explaining the use of metaverse technology among students attending higher education institutions in Egypt, the research findings suggested that autonomy and perceived usefulness (PU) are major determinants. Nevertheless, the continuing intention was unaffected by the perceived ease of use (PEOU) of the product. Furthermore, according to the data provided by the ANN model, the most significant predictors are relatedness, PEOU, autonomy, and PU. It has been determined that the results obtained from the PLS-SEM and ANN modes are identical. Additionally, both theoretical and practical implications are discussed in this article.

Testing the Cosmic Distance Duality Relation Using Strong Gravitational Lensing Time Delays and Type Ia Supernovae

Abstract We present a comprehensive test of the cosmic distance duality relation (DDR) using a combination of strong gravitational lensing (SGL) time delay measurements and Type Ia supernovae (SNe Ia) data. We investigate three different parameterizations of potential DDR violations. To bridge the gap between SGL and SNe Ia data sets, we implement an artificial neural network approach to reconstruct the distance modulus of SNe Ia. Our analysis uniquely considers both scenarios where the absolute magnitude of SNe Ia (M B ) is treated as a free parameter and where it is fixed to a Cepheid-calibrated value. Using a sample of six SGL systems and the Pantheon+ SNe Ia data set, we find no statistically significant evidence for DDR violations across all parameterizations. The consistency of our findings across different parameterizations not only reinforces confidence in the standard DDR but also demonstrates the robustness of our analytical approach.

A multilevel social network approach to studying multiple disease-prevention behaviors

The effective prevention of many infectious and non-infectious diseases relies on people concurrently adopting multiple prevention behaviors. Individual characteristics, opinion leaders, and social networks have been found to explain why people take up specific prevention behaviors. However, it remains challenging to understand how these factors shape multiple interdependent behaviors. We propose a multilevel social network framework that allows us to study the effects of individual and social factors on multiple disease prevention behaviors simultaneously. We apply this approach to examine the factors explaining eight malaria prevention behaviors, using unique interview data collected from 1529 individuals in 10 hard-to-reach, malaria-endemic villages in Meghalaya, India in 2020–2022. Statistical network modelling reveals exposure to similar behaviors in one’s social network as the most important factor explaining prevention behaviors. Further, we find that households indirectly shape behaviors as key contexts for social ties. Together, these two factors are crucial for explaining the observed patterns of behaviors and social networks in the data, outweighing individual characteristics, opinion leaders, and social network size. The results highlight that social network processes may facilitate or hamper disease prevention efforts that rely on a combination of behaviors. Our approach is well suited to study these processes in the context of various diseases.

DSF-GAM: a location inference model in social network Twitter

Analyzing the social interactions and texts on Twitter can provide valuable insights into users' behavior, opinions, and even their geographical locations. Location inference of users within a social network finds applications in various domains, such as targeted advertising and public health monitoring. Inference techniques through social interactions, such as co-mentions, remain a favored method for constructing user social networks. Co-mentions, however, introduce a substantial number of noisy relationships that impact the overall network structure and prediction of the location of a user. Furthermore, previous approaches have overlooked the examination of cohesive groups or community structures within an individual's ego network for location prediction. This paper introduces a predictive model called DSF-GAM (DS stand for document similarity, F stands for frequencies and GAM stands for Generalized Additive Model). which redesigns social interactions based on mentions to infer user location on Twitter. DSF-GAM introduces a novel and innovative social graph, termed a user relation graph. DSF-GAM employs a generalized additive model (GAM) for location inference. The fundamental contribution of DSF-GAM lies in the representation of data using a new graph structure with enriched community features and edges. The experimental results on well-known datasets demonstrate that DSF-GAM outperforms other state-of-the-art methods based on both content and network approaches and mixed results in hybrid approaches.

Decision Tree Model-Based Security Situational Awareness Approach for Wireless Communication Networks

Network security situational awareness is gaining increasing attention due to its capability to globally and dynamically detect potential network security risks. However, traditional security situational awareness models often exhibit poor classification performance, resulting in lower-than-expected acceleration and scalability ratios. In this paper, we propose a novel security situational awareness approach for wireless communication networks based on a decision tree model. First, reconfigure the category division module to categorize the attack data into four different types. Then, using time windows to segment the data flow between the network and the host promotes the design of effective security event detection mechanisms in the model. Finally, a comprehensive network security situational awareness model was constructed at the joint level using decision tree algorithm. The experimental results show that the proposed method can significantly improve the acceleration ratio, and the space occupancy ratio can reach 80, indicating that the proposed method can have a high level of processing capability and accurate perception in network security situations, providing a guarantee for the security of wireless communication networks.

A novel muscle network approach for objective assessment and profiling of bulbar involvement in ALS

A novel muscle network approach for objective assessment and profiling of bulbar involvement in ALS

Passive Monitoring of Parkinson Tremor in Daily Life: A Prototypical Network Approach

Objective and continuous monitoring of Parkinson’s disease (PD) tremor in free-living conditions could benefit both individual patient care and clinical trials, by overcoming the snapshot nature of clinical assessments. To enable robust detection of tremor in the context of limited amounts of labeled training data, we propose to use prototypical networks, which can embed domain expertise about the heterogeneous tremor and non-tremor sub-classes. We evaluated our approach using data from the Parkinson@Home Validation study, including 8 PD patients with tremor, 16 PD patients without tremor, and 24 age-matched controls. We used wrist accelerometer data and synchronous expert video annotations for the presence of tremor, captured during unscripted daily life activities in and around the participants’ own homes. Based on leave-one-subject-out cross-validation, we demonstrate the ability of prototypical networks to capture free-living tremor episodes. Specifically, we demonstrate that prototypical networks can be used to enforce robust performance across domain-informed sub-classes, including different tremor phenotypes and daily life activities.

Effect of Inter Turbine Spacing in Omnidirectional Wind for Savonius Cluster: A Computational Fluid Dynamics and Artificial Neural Network Approach

Effect of Inter Turbine Spacing in Omnidirectional Wind for Savonius Cluster: A Computational Fluid Dynamics and Artificial Neural Network Approach