Data-driven Computational Approaches Research Articles

Data-driven computational approaches to estimate gross calorific value of coal using proximate and ultimate analyses

ABSTRACT A crucial factor in determining coal’s quality is its gross calorific value (GCV). GCV is generally determined experimentally using an adiabatic bomb calorimeter. But the laboratory measurement process is complicated, tedious, expensive, and time-consuming. To address these problems, this study has developed five machine learning techniques, such as multilayer perceptron artificial neural network (MLP-ANN), multiple variable regression (MVR), adaptive neuro-fuzzy inference system (ANFIS), and support vector machine (SVM). ANN is further regularized using early stopping to avoid overfitting, whereas the MVR is regularized by employing L1, L2, and elastic networks. An inter-item correlation matrix, scatterplots, and sensitivity analysis are used to rank the parameters of the ultimate and proximate analyses according to their significance. A total of 6582 datapoints from the US. Geological Survey Coal Quality (COALQUAL) database are utilized for this study. The models’ performance is measured using coefficient of determinant (), mean squared error (MSE), explained error (EV), maximum error (max_error), minimum error (min_error), mean absolute error (MAE), and mean absolute percentage error (MAPE), respectively. The study’s findings demonstrate that MVR, MLP-ANN, SVM, and ANFIS are highly efficient concerning both prediction and computation effectiveness. The prediction performances of these models are somewhat similar, with a coefficient of determinant of over 0.99. Hence, simple MVR is sufficient to predict GCV accurately instead of complex modeling. Parameters ranking indicates that oxygen (O), carbon (C), and moisture (M) are the most significant input variables for GCV prediction.

Graph-based representation of history-dependent material response in the Data-Driven Computational Mechanics framework

This work develops an extension of the Data-Driven Computational Mechanics approach introduced by Kirchdoerfer and Ortiz (2016) to history-dependent behaviour. The material database is replaced by a material directed graph, where vertices represent strain–stress pairs and arcs encode thermomechanically admissible transitions between them. A graph search algorithm is used to select local material databases representative of the incremental material behaviour. This study thus gives the mathematical tools for a structured data-driven representation of the material behaviour, designed to minimise the increase of the problem dimensionality. The performances of the proposed approach applied to elastoplasticity are studied on a single-element structure and a larger truss structure. A two-step solver is proposed to address numerical accuracy issues in the latter case.

Systematic transcriptome-wide meta-analysis across endocrine disrupting chemicals reveals shared and unique liver pathways, gene networks, and disease associations

Cardiometabolic disorders (CMD) are a growing public health problem across the world. Among the known cardiometabolic risk factors are compounds that induce endocrine and metabolic dysfunctions, such as endocrine disrupting chemicals (EDCs). To date, how EDCs influence molecular programs and cardiometabolic risks has yet to be fully elucidated, especially considering the complexity contributed by species-, chemical-, and dose-specific effects. Moreover, different experimental and analytical methodologies employed by different studies pose challenges when comparing findings across studies. To explore the molecular mechanisms of EDCs in a systematic manner, we established a data-driven computational approach to meta-analyze 30 human, mouse, and rat liver transcriptomic datasets for 4 EDCs, namely bisphenol A (BPA), bis(2-ethylhexyl) phthalate (DEHP), tributyltin (TBT), and perfluorooctanoic acid (PFOA). Our computational pipeline uniformly re-analyzed pre-processed quality-controlled microarray data and raw RNAseq data, derived differentially expressed genes (DEGs) and biological pathways, modeled gene regulatory networks and regulators, and determined CMD associations based on gene overlap analysis. Our approach revealed that DEHP and PFOA shared stable transcriptomic signatures that are enriched for genes associated with CMDs, suggesting similar mechanisms of action such as perturbations of peroxisome proliferator-activated receptor gamma (PPARγ) signaling and liver gene network regulators VNN1 and ACOT2. In contrast, TBT exhibited highly divergent gene signatures, pathways, network regulators, and disease associations from the other EDCs. In addition, we found that the rat, mouse, and human BPA studies showed highly variable transcriptomic patterns, providing molecular support for the variability in BPA responses. Our work offers insights into the commonality and differences in the molecular mechanisms of various EDCs and establishes a streamlined data-driven workflow to compare molecular mechanisms of environmental substances to elucidate the underlying connections between chemical exposure and disease risks.

Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium

BackgroundRecent advances in data-driven computational approaches have been helpful in devising tools to objectively diagnose psychiatric disorders. However, current machine learning studies limited to small homogeneous samples, different methodologies, and different imaging collection protocols, limit the ability to directly compare and generalize their results. Here we aimed to classify individuals with PTSD versus controls and assess the generalizability using a large heterogeneous brain datasets from the ENIGMA-PGC PTSD Working group. MethodsWe analyzed brain MRI data from 3,477 structural-MRI; 2,495 resting state-fMRI; and 1,952 diffusion-MRI. First, we identified the brain features that best distinguish individuals with PTSD from controls using traditional machine learning methods. Second, we assessed the utility of the denoising variational autoencoder (DVAE) and evaluated its classification performance. Third, we assessed the generalizability and reproducibility of both models using leave-one-site-out cross-validation procedure for each modality. ResultsWe found lower performance in classifying PTSD vs. controls with data from over 20 sites (60 % test AUC for s-MRI, 59 % for rs-fMRI and 56 % for d-MRI), as compared to other studies run on single-site data. The performance increased when classifying PTSD from HC without trauma history in each modality (75 % AUC). The classification performance remained intact when applying the DVAE framework, which reduced the number of features. Finally, we found that the DVAE framework achieved better generalization to unseen datasets compared with the traditional machine learning frameworks, albeit performance was slightly above chance. ConclusionThese results have the potential to provide a baseline classification performance for PTSD when using large scale neuroimaging datasets. Our findings show that the control group used can heavily affect classification performance. The DVAE framework provided better generalizability for the multi-site data. This may be more significant in clinical practice since the neuroimaging-based diagnostic DVAE classification models are much less site-specific, rendering them more generalizable.

Data Driven Discovery of MOFs for Hydrogen Gas Adsorption.

Hydrogen gas (H2) is a clean and renewable energy source, but the lack of efficient and cost-effective storage materials is a challenge to its widespread use. Metal-organic frameworks (MOFs), a class of porous materials, have been extensively studied for H2 storage due to their tunable structural and chemical features. However, the large design space offered by MOFs makes it challenging to select or design appropriate MOFs with a high H2 storage capacity. To overcome these challenges, we present a data-driven computational approach that systematically designs new functionalized MOFs for H2 storage. In particular, we showcase the framework of a hybrid particle swarm optimization integrated genetic algorithm, grand canonical Monte Carlo (GCMC) simulations, and our in-house MOF structure generation code to design new MOFs with excellent H2 uptake. This automated, data driven framework adds appropriate functional groups to IRMOF-10 to improve its H2 adsorption capacity. A detailed analysis of the top selected MOFs, their adsorption isotherms, and MOF design rules to enhance H2 adsorption are presented. We found a functionalized IRMOF-10 with an enhanced H2 adsorption increased by ∼6 times compared to that of pure IRMOF-10 at 1 bar and 77 K. Furthermore, this study also utilizes machine learning and deep learning techniques to analyze a large data set of MOF structures and properties, in order to identify the key factors that influence hydrogen adsorption. The proof-of-concept that uses a machine learning/deep learning approach to predict hydrogen adsorption based on the identified structural and chemical properties of the MOF is demonstrated.

A review of neuroimaging-based data-driven approach for Alzheimer's disease heterogeneity analysis.

Alzheimer's disease (AD) is a complex form of dementia and due to its high phenotypic variability, its diagnosis and monitoring can be quite challenging. Biomarkers play a crucial role in AD diagnosis and monitoring, but interpreting these biomarkers can be problematic due totheir spatial and temporal heterogeneity. Therefore, researchers are increasingly turning to imaging-based biomarkers that employ data-driven computational approaches to examine the heterogeneity of AD. In this comprehensive review article, we aim to provide health professionals with a comprehensive view of past applications of data-driven computational approaches in studying AD heterogeneity and planning future research directions. We first define and offer basic insights into different categories of heterogeneity analysis, including spatial heterogeneity, temporal heterogeneity, and spatial-temporal heterogeneity. Then, we scrutinize 22 articles relating to spatial heterogeneity, 14 articles relating to temporal heterogeneity, and five articles relating to spatial-temporal heterogeneity, highlighting the strengths and limitations of these strategies. Furthermore, we discuss the importance of understanding spatial heterogeneity in AD subtypes and their clinical manifestations, biomarkers for abnormal orderings and AD stages, the recent advancements in spatial-temporal heterogeneity analysis for AD, and the emerging role of omics data integration in advancing personalized diagnosis and treatment for AD patients. By emphasizing the significance of understanding AD heterogeneity, we hope to stimulate further research in this field to facilitate the development of personalized interventions for AD patients.

Trustworthiness judgments without the halo effect: A data-driven computational modeling approach.

Trustworthy-looking faces are also perceived as more attractive, but are there other meaningful cues that contribute to perceived trustworthiness? Using data-driven models, we identify these cues after removing attractiveness cues. In Experiment 1, we show that both judgments of trustworthiness and attractiveness of faces manipulated by a model of perceived trustworthiness change in the same direction. To control for the effect of attractiveness, we build two new models of perceived trustworthiness: a subtraction model, which forces the perceived attractiveness and trustworthiness to be negatively correlated (Experiment 2), and an orthogonal model, which reduces their correlation (Experiment 3). In both experiments, faces manipulated to appear more trustworthy were indeed perceived to be more trustworthy, but not more attractive. Importantly, in both experiments, these faces were also perceived as more approachable and with more positive expressions, as indicated by both judgments and machine learning algorithms. The current studies show that the visual cues used for trustworthiness and attractiveness judgments can be separated, and that apparent approachability and facial emotion are driving trustworthiness judgments and possibly general valence evaluation.

A machine learning-based probabilistic computational framework for uncertainty quantification of actuation of clustered tensegrity structures.

Clustered tensegrity structures integrated with continuous cables are lightweight, foldable, and deployable. Thus, they can be used as flexible manipulators or soft robots. The actuation process of such soft structure has high probabilistic sensitivity. It is essential to quantify the uncertainty of actuated responses of the tensegrity structures and to modulate their deformation accurately. In this work, we propose a comprehensive data-driven computational approach to study the uncertainty quantification (UQ) and probability propagation in clustered tensegrity structures, and we have developed a surrogate optimization model to control the flexible structure deformation. An example of clustered tensegrity beam subjected to a clustered actuation is presented to demonstrate the validity of the approach and its potential application. The three main novelties of the data-driven framework are: (1) The proposed model can avoid the difficulty of convergence in nonlinear Finite Element Analysis (FEA), by two machine learning methods, the Gauss Process Regression (GPR) and Neutral Network (NN). (2) A fast real-time prediction on uncertainty propagation can be achieved by the surrogate model, and (3) Optimization of the actuated deformation comes true by using both Sequence Quadratic Programming (SQP) and Bayesian optimization methods. The results have shown that the proposed data-driven computational approach is powerful and can be extended to other UQ models or alternative optimization objectives.

Predicting future well performance for environmental remediation design using deep learning

In this study, we developed a deep learning (DL) framework with a multi-channel three-dimensional convolutional neural network (MC3D-CNN) to predict well performance and thereby assist future environmental remediation design. Such prediction of extraction well performance at designated locations is critical for configuring pump-and-treat (P&T) well network design and operation, setting reasonable target closure dates for overall remedying, and estimating remedy costs. The framework is developed with operational and monitoring data routinely collected during P&T remedy operations, including well extraction and injection rates as well as in situ contaminant concentrations. Traditionally, the collected data were rarely used for purposes other than assessing past well performance and the accuracy of the conceptual site model. However, recent advances in data-driven computational approaches enable better use of the large datasets to inform future well performance, enhance site characterization, and improve remediation planning. In this study, we established a DL framework to integrate transient three-dimensional contaminant plumes and multiple aquifer properties (e.g., hydraulic conductivity and hydrostratigraphic maps) to identify characteristic patterns controlling and representing extraction well mass recovery, aiming at providing future mass recovery estimates for existing wells and candidate wells at any proposed locations. We evaluated our framework by using a realistic synthetic dataset generated from a well-calibrated flow and transport model used in the 200 West Area of the U.S. Department of Energy’s Hanford Site in southeastern Washington state. The multi-channel feature in our framework allows integration of various types and temporal densities of training datasets for DL model development. Overall, we found that the trained DL model achieved an accuracy of over 90% in ranking extraction well performance in validation datasets, and over 80% in predicting high-performance-ranking well locations. This data-informed approach provides a flexible tool to support adaptive site management, streamline decision-making, and potentially reduce remediation time and costs. Our DL framework can be used as a filtering tool to improve the current P&T network optimization design by reducing the number of candidate well locations.

Computational approach for structure generation of anisotropic particles (CASGAP) with targeted distributions of particle design and orientational order.

The macroscopic properties of materials are governed by their microscopic structure which depends on the materials' composition (i.e., building blocks) and processing conditions. In many classes of synthetic, bioinspired, or natural soft and/or nanomaterials, one can find structural anisotropy in the microscopic structure due to anisotropic building blocks and/or anisotropic domains formed through the processing conditions. Experimental characterization and complementary physics-based or data-driven modeling of materials' structural anisotropy are critical for understanding structure-property relationships and enabling targeted design of materials with desired macroscopic properties. In this pursuit, to interpret experimentally obtained characterization results (e.g., scattering profiles) of soft materials with structural anisotropy using data-driven computational approaches, there is a need for creating real space three-dimensional structures of the designer soft materials with realistic physical features (e.g., dispersity in building block sizes) and anisotropy (i.e., aspect ratios of the building blocks, their orientational and positional order). These real space structures can then be used to compute and complement experimentally obtained characterization results or be used as initial configurations for physics-based simulations/calculations that can then provide training data for machine learning models. To address this need, we present a new computational approach called CASGAP - Computational Approach for Structure Generation of Anisotropic Particles - for generating any desired three dimensional real-space structure of anisotropic building blocks (modeled as particles) adhering to target distributions of particle shape, size, and positional and orientational order.

Drug Repurposing at the Interface of Melanoma Immunotherapy and Autoimmune Disease.

Cancer cells have a remarkable ability to evade recognition and destruction by the immune system. At the same time, cancer has been associated with chronic inflammation, while certain autoimmune diseases predispose to the development of neoplasia. Although cancer immunotherapy has revolutionized antitumor treatment, immune-related toxicities and adverse events detract from the clinical utility of even the most advanced drugs, especially in patients with both, metastatic cancer and pre-existing autoimmune diseases. Here, the combination of multi-omics, data-driven computational approaches with the application of network concepts enables in-depth analyses of the dynamic links between cancer, autoimmune diseases, and drugs. In this review, we focus on molecular and epigenetic metastasis-related processes within cancer cells and the immune microenvironment. With melanoma as a model, we uncover vulnerabilities for drug development to control cancer progression and immune responses. Thereby, drug repurposing allows taking advantage of existing safety profiles and established pharmacokinetic properties of approved agents. These procedures promise faster access and optimal management for cancer treatment. Together, these approaches provide new disease-based and data-driven opportunities for the prediction and application of targeted and clinically used drugs at the interface of immune-mediated diseases and cancer towards next-generation immunotherapies.

Interconnectedness and (in)coherence as a signature of conspiracy worldviews.

Conspiracy theories may arise out of an overarching conspiracy worldview that identifies common elements of subterfuge across unrelated or even contradictory explanations, leading to networks of self-reinforcing beliefs. We test this conjecture by analyzing a large natural language database of conspiracy and nonconspiracy texts for the same events, thus linking theory-driven psychological research with data-driven computational approaches. We find that, relative to nonconspiracy texts, conspiracy texts are more interconnected, more topically heterogeneous, and more similar to one another, revealing lower cohesion within texts but higher cohesion between texts and providing strong empirical support for an overarching conspiracy worldview. Our results provide inroads for classification algorithms and further exploration into individual differences in belief structures.

Representative random sampling: an empirical evaluation of a novel bin stratification method for model performance estimation

High-dimensional cancer data can be burdensome to analyze, with complex relationships between molecular measurements, clinical diagnostics, and treatment outcomes. Data-driven computational approaches may be key to identifying relationships with potential clinical or research use. To this end, reliable comparison of feature engineering approaches in their ability to support machine learning survival modeling is crucial. With the limited number of cases often present in multi-omics datasets (“big p, little n,” or many features, few subjects), a resampling approach such as cross validation (CV) would provide robust model performance estimates at the cost of flexibility in intermediate assessments and exploration in feature engineering approaches. A holdout (HO) estimation approach, however, would permit this flexibility at the expense of reliability. To provide more reliable HO-based model performance estimates, we propose a novel sampling procedure: representative random sampling (RRS). RRS is a special case of continuous bin stratification which minimizes significant relationships between random HO groupings (or CV folds) and a continuous outcome. Monte Carlo simulations used to evaluate RRS on synthetic molecular data indicated that RRS-based HO (RRHO) yields statistically significant reductions in error and bias when compared with standard HO. Similarly, more consistent reductions are observed with RRS-based CV. While resampling approaches are the ideal choice for performance estimation with limited data, RRHO can enable more reliable exploratory feature engineering than standard HO.

CSM-peptides: A computational approach to rapid identification of therapeutic peptides.

Peptides are attractive alternatives for the development of new therapeutic strategies due to their versatility and low complexity of synthesis. Increasing interest in these molecules has led to the creation of large collections of experimentally characterized therapeutic peptides, which greatly contributes to development of data‐driven computational approaches. Here we propose CSM‐peptides, a novel machine learning method for rapid identification of eight different types of therapeutic peptides: anti‐angiogenic, anti‐bacterial, anti‐cancer, anti‐inflammatory, anti‐viral, cell‐penetrating, quorum sensing, and surface binding. Our method has shown to outperform existing approaches, achieving an AUC of up to 0.92 on independent blind tests, and consistent performance on cross‐validation. We anticipate CSM‐peptides to be of great value in helping screening large libraries to identify novel peptides with therapeutic potential and have made it freely available as a user‐friendly web server and Application Programming Interface at https://biosig.lab.uq.edu.au/csm_peptides.

Doctor trustworthiness influences pain and its neural correlates in virtual medical interactions.

Trust is an important component of the doctor-patient relationship and is associated with improved patient satisfaction and health outcomes. Previously, we reported that patient feelings of trust and similarity toward their clinician predicted reductions in evoked pain in response to painful heat stimulations. In the present study, we investigated the brain mechanisms underlying this effect. We used face stimuli previously developed using a data-driven computational modeling approach that differ in perceived trustworthiness and superimposed them on bodies dressed in doctors' attire. During functional magnetic resonance imaging, participants (n = 42) underwent a series of virtual medical interactions with these doctors during which they received painful heat stimulation as an analogue of a painful diagnostic procedure. Participants reported increased pain when receiving painful heat stimulations from low-trust doctors, which was accompanied by increased activity in pain-related brain regions and a multivariate pain-predictive neuromarker. Findings suggest that patient trust in their doctor may have tangible impacts on pain and point to a potential brain basis for trust-related reductions in pain through the modulation of brain circuitry associated with the sensory-discriminative and affective-motivational dimensions of pain.

Emotion Recognition in a Multi-Componential Framework: The Role of Physiology

The Component Process Model is a well-established framework describing an emotion as a dynamic process with five highly interrelated components: cognitive appraisal, expression, motivation, physiology and feeling. Yet, few empirical studies have systematically investigated discrete emotions through this full multi-componential view. We therefore elicited various emotions during movie watching and measured their manifestations across these components. Our goal was to investigate the relationship between physiological measures and the theoretically defined components, as well as to determine whether discrete emotions could be predicted from the multicomponent response patterns. By deploying a data-driven computational approach based on multivariate pattern classification, our results suggest that physiological features are encoded within each component, supporting the hypothesis of a synchronized recruitment during an emotion episode. Overall, while emotion prediction was higher when classifiers were trained with all five components, a model without physiology features did not significantly reduce the performance. The findings therefore support a description of emotion as a multicomponent process, in which emotion recognition requires the integration of all the components. However, they also indicate that physiologyper seis the least significant predictor for emotion classification among these five components.

Reduced inhibition in depression impairs stimulus processing in human cortical microcircuits.

Cortical processing depends on finely tuned excitatory and inhibitory connections in neuronal microcircuits. Reduced inhibition by somatostatin-expressing interneurons is a key component of altered inhibition associated with treatment-resistant major depressive disorder (depression), which is implicated in cognitive deficits and rumination, but the link remains to be better established mechanistically in humans. Here we test the effect of reduced somatostatin interneuron-mediated inhibition on cortical processing in human neuronal microcircuits using a data-driven computational approach. We integrate human cellular, circuit, and gene expression data to generate detailed models of human cortical microcircuits in health and depression. We simulate microcircuit baseline and response activity and find a reduced signal-to-noise ratio and increased false/failed detection of stimuli due to a higher baseline activity in depression. We thus apply models of human cortical microcircuits to demonstrate mechanistically how reduced inhibition impairs cortical processing in depression, providing quantitative links between altered inhibition and cognitive deficits.

Computationally prioritized drugs inhibit SARS-CoV-2 infection and syncytia formation

The pharmacological arsenal against the COVID-19 pandemic is largely based on generic anti-inflammatory strategies or poorly scalable solutions. Moreover, as the ongoing vaccination campaign is rolling slower than wished, affordable and effective therapeutics are needed. To this end, there is increasing attention toward computational methods for drug repositioning and de novo drug design.Here, multiple data-driven computational approaches are systematically integrated to perform a virtual screening and prioritize candidate drugs for the treatment of COVID-19. From the list of prioritized drugs, a subset of representative candidates to test in human cells is selected. Two compounds, 7-hydroxystaurosporine and bafetinib, show synergistic antiviral effects in vitro and strongly inhibit viral-induced syncytia formation. Moreover, since existing drug repositioning methods provide limited usable information for de novo drug design, the relevant chemical substructures of the identified drugs are extracted to provide a chemical vocabulary that may help to design new effective drugs.

Data-driven Exploration of Engagement with Workplace-based Assessment in the Clinical Skills Domain

The paper presents a multi-faceted data-driven computational approach to analyse workplace-based assessment (WBA) of clinical skills in medical education. Unlike formal university-based part of the degree, the setting of WBA can be informal and only loosely regulated, as students are encouraged to take every opportunity to learn from the clinical setting. For clinical educators and placement coordinators it is vital to follow and analyse students’ engagement with WBA while on placements, in order to understand how students are participating in the assessment, and what improvements can be made. We analyse digital data capturing the students’ WBA attempts and comments on how the assessments went, using process mining and text analytics. We compare Year 1 cohorts across three years, focusing on differences between primary vs. secondary care placements. The main contribution of the work presented in this paper is the exploration of computational approaches for multi-faceted, data-driven assessment analytics for workplace learning which includes:(i) a set of features for analysing clinical skills WBA data, (ii) analysis of the temporal aspects ofthat data using process mining, and (iii) utilising text analytics to compare student reflections on WBA. We show how assessment data captured during clinical placements can provide insights about the student engagement and inform the medical education practice. Our work is inspired by Jim Greer’s vision that intelligent methods and techniques should be adopted to address key challenges faced by educational practitioners in order to foster improvement of learning and teaching. In the broader AI in Education context, the paper shows the application of AI methods to address educational challenges in a new informal learning domain - practical healthcare placements in higher education medical training.

Parsing the Mechanistic Heterogeneity of Hallucinations Using Theory- and Data-Driven Computational Approaches

Behavioral task performance can be driven by latent factors which are not directly observed. Generative models of performance can reveal those latent factors, which may themselves help us understand the systems driving performance. Computational models of perception could identify the latent factors underlying hallucinations. Modeling of performance on an auditory conditioned hallucinations(ACH) task shows that hallucinations may arise from an over-weighting of prior beliefs during perception. However, it remains unclear whether this susceptibility reflects hallucination state or trait vulnerability, is specific to perceptual modality, and whether it is present in all types of hallucinations.