Computational Model Research Articles

A stochastic world model on gravity for stability inference.

The fact that objects without proper support will fall to the ground is not only a natural phenomenon, but also common sense in mind. Previous studies suggest that humans may infer objects' stability through a world model that performs mental simulations with a priori knowledge of gravity acting upon the objects. Here we measured participants' sensitivity to gravity to investigate how the world model works. We found that the world model on gravity was not a faithful replica of the physical laws, but instead encoded gravity's vertical direction as a Gaussian distribution. The world model with this stochastic feature fit nicely with participants' subjective sense of objects' stability and explained the illusion that taller objects are perceived as more likely to fall. Furthermore, a computational model with reinforcement learning revealed that the stochastic characteristic likely originated from experience-dependent comparisons between predictions formed by internal simulations and the realities observed in the external world, which illustrated the ecological advantage of stochastic representation in balancing accuracy and speed for efficient stability inference. The stochastic world model on gravity provides an example of how a priori knowledge of the physical world is implemented in mind that helps humans operate flexibly in open-ended environments.

Left Ventricular Unloading in Extracorporeal Membrane Oxygenation: A Clinical Perspective Derived from Basic Cardiovascular Physiology.

To present an abridged overview of the literature and pathophysiological background of adjunct interventional left ventricular unloading strategies during veno-arterial extracorporeal membrane oxygenation (V-A ECMO). From a clinical perspective, the mechanistic complexity of such combined mechanical circulatory support often requires in-depth physiological reasoning at the bedside, which remains a cornerstone of daily practice for optimal patient-specific V-A ECMO care. Recent conventional clinical trials have not convincingly shown the superiority of V-A ECMO in acute myocardial infarction complicated by cardiogenic shock as compared with medical therapy alone. Though, it has repeatedly been reported that the addition of interventional left ventricular unloading to V-A ECMO may improve clinical outcome. Novel approaches such as registry-based adaptive platform trials and computational physiological modeling are now introduced to inform clinicians by aiming to better account for patient-specific variation and complexity inherent to V-A ECMO and have raised a widespread interest. To provide modern high-quality V-A ECMO care, it remains essential to understand the patient's pathophysiology and the intricate interaction of an individual patient with extracorporeal circulatory support devices. Innovative clinical trial design and computational modeling approaches carry great potential towards advanced clinical decision support in ECMO and related critical care.

Application of digital twins for simulation based tailoring of laser induced graphene

In the era of man–machine interfaces, digital twins stand as a key technology, offering virtual representations of real-world objects, processes, and systems through computational models. They enable novel ways of interacting with, comprehending, and manipulating real-world entities within a virtual realm. The real implementation of graphene-based sensors and electronic devices remains challenging due to the integration complexities of high-quality graphene materials with existing manufacturing processes. To address this, scalable techniques for the in-situ fabrication of graphene-like materials are essential. One promising method involves using a CO2 laser to convert polyimide into graphene. Optimizing this graphitization process is hindered by complex parameter interactions and nonlinear terms. This article explores how these digital replicas can enhance the fabrication of laser-induced graphene (LIG) through laser simulation and machine learning methods to enable rapid single-step LIG patterning. This approach aims to create a universal simulation for all CO2 lasers, calculating optical energy flux and utilizing machine learning to control and predict LIG conductivity (ability to conduct current), morphology, and electrical resistance. The proposed procedure, integrating digital twins in the LIG production process, will avoid or reduce the preliminary tests required to determine the proper laser parameters to reach the desired LIG characteristics. Accordingly, this approach will reduce the time and costs associated with these tests and thus increase the efficiency and optimize the procedure.

Identification and characterization of a new potent inhibitor targeting CtBP1/BARS in melanoma cells

BackgroundThe C-terminal-binding protein 1/brefeldin A ADP-ribosylation substrate (CtBP1/BARS) acts both as an oncogenic transcriptional co-repressor and as a fission inducing protein required for membrane trafficking and Golgi complex partitioning during mitosis, hence for mitotic entry. CtBP1/BARS overexpression, in multiple cancers, has pro-tumorigenic functions regulating gene networks associated with “cancer hallmarks” and malignant behavior including: increased cell survival, proliferation, migration/invasion, epithelial-mesenchymal transition (EMT).Structurally, CtBP1/BARS belongs to the hydroxyacid-dehydrogenase family and possesses a NAD(H)-binding Rossmann fold, which, depending on ligands bound, controls the oligomerization of CtBP1/BARS and, in turn, its cellular functions.Here, we proposed to target the CtBP1/BARS Rossmann fold with small molecules as selective inhibitors of mitotic entry and pro-tumoral transcriptional activities.MethodsStructured-based screening of drug databases at different development stages was applied to discover novel ligands targeting the Rossmann fold. Among these identified ligands, N-(3,4-dichlorophenyl)-4-{[(4-nitrophenyl)carbamoyl]amino}benzenesulfonamide, called Comp.11, was selected for further analysis. Fluorescence spectroscopy, isothermal calorimetry, computational modelling and site-directed mutagenesis were employed to define the binding of Comp.11 to the Rossmann fold. Effects of Comp.11 on the oligomerization state, protein partners binding and pro-tumoral activities were evaluated by size-exclusion chromatography, pull-down, membrane transport and mitotic entry assays, Flow cytometry, quantitative real-time PCR, motility/invasion, and colony assays in A375MM and B16F10 melanoma cell lines. Effects of Comp.11 on tumor growth in vivo were analyzed in mouse tumor model.ResultsWe identify Comp.11 as a new, potent and selective inhibitor of CtBP1/BARS (but not CtBP2). Comp.11 directly binds to the CtBP1/BARS Rossmann fold affecting the oligomerization state of the protein (unlike other known CtBPs inhibitors), which, in turn, hinders interactions with relevant partners, resulting in the inhibition of both CtBP1/BARS cellular functions: i) membrane fission, with block of mitotic entry and cellular secretion; and ii) transcriptional pro-tumoral effects with significantly hampered proliferation, EMT, migration/invasion, and colony-forming capabilities. The combination of these effects impairs melanoma tumor growth in mouse models. ConclusionsThis study identifies a potent and selective inhibitor of CtBP1/BARS active in cellular and melanoma animal models revealing new opportunities to study the role of CtBP1/BARS in tumor biology and to develop novel melanoma treatments.

Removing direct photocurrent artifacts in optogenetic connectivity mapping data via constrained matrix factorization.

Monosynaptic connectivity mapping is crucial for building circuit-level models of neural computation. Two-photon optogenetic stimulation, when combined with whole-cell recording, enables large-scale mapping of physiological circuit parameters. In this experimental setup, recorded postsynaptic currents are used to infer the presence and strength of connections. For many cell types, nearby connections are those we expect to be strongest. However, when the postsynaptic cell expresses opsin, optical excitation of nearby cells can induce direct photocurrents in the postsynaptic cell. These photocurrent artifacts contaminate synaptic currents, making it difficult or impossible to probe connectivity for nearby cells. To overcome this problem, we developed a computational tool, Photocurrent Removal with Constraints (PhoRC). Our method is based on a constrained matrix factorization model which leverages the fact that photocurrent kinetics are less variable than those of synaptic currents. We demonstrate on real and simulated data that PhoRC consistently removes photocurrents while preserving synaptic currents, despite variations in photocurrent kinetics across datasets. Our method allows the discovery of synaptic connections which would have been otherwise obscured by photocurrent artifacts, and may thus reveal a more complete picture of synaptic connectivity. PhoRC runs faster than real time and is available as open source software.

A model with multiple intracranial aneurysms: possible hemodynamic mechanisms of aneurysmal initiation, rupture and recurrence

BackgroundHemodynamic factors play an important role in aneurysm initiation, growth, rupture, and recurrence, while the mechanism of the hemodynamic characteristics is still controversial. A unique model of multiple aneurysms (initiation, growth, rupture, and recurrence) is helpful to avoids the confounders and further explore the possible hemodynamic mechanisms of aneurysm in different states.MethodsWe present a model with multiple aneurysms, and including the states of initiation, growth, rupture, and recurrence, discuss the proposed mechanisms, and describe computational fluid dynamic model that was used to evaluate the likely hemodynamic effect of different states of the aneurysms.ResultsThe hemodynamic analysis suggests that high flow impingement and high WSS distribution at normal parent artery was found before aneurysmal initiation. The WSS distribution and flow velocity were decreased in the new sac after aneurysmal growth. Low WSS was the risk hemodynamic factor for aneurysmal rupture. High flow concentration region on the neck plane after coil embolization still marked in recanalized aneurysm.ConclusionsAssociations have been identified between high flow impingement and aneurysm recanalization, while low WSS is linked to the rupture of aneurysms. High flow concentration and high WSS distribution at normal artery associated with aneurysm initiation and growth, while after growth, the high-risk hemodynamics of aneurysm rupture was occurred, which is low WSS at aneurysm dome.

In-Depth Analysis of the Selection of PBPK Modeling Tools: Bibliometric and Social Network Analysis of the Open Systems Pharmacology Community.

Since the Open Source Initiative laid the foundation for the open source software environment in 1998, the popularity of free and open source software has been steadily increasing. Model-informed drug discovery and development (MID3), a key component of pharmaceutical research and development, heavily makes use of computational models which can be developed using various software including the Open Systems Pharmacology (OSP) software (PK-Sim/MoBi), a free and open source software tool for physiologically based pharmacokinetic (PBPK) modeling. In this study, we aimed to investigate the impact, application areas, and reach of the OSP software as well as the relationships and collaboration patterns between organizations having published OSP-related articles between 2017 and 2023. Therefore, we conducted a bibliometric analysis of OSP-related publications and a social network analysis of the organizations with which authors of OSP-related publications were affiliated. On several levels, we found evidence for a significant growth in the size of the OSP community as well as its visibility in the MID3 community since OSP's establishment in 2017. Specifically, the annual publication rate of PubMed-indexed PBPK-related articles using the OSP software outpaced that of PBPK-related articles using any software. Our bibliometric analysis and network analysis demonstrated that the expansion of the OSP community was predominantly driven by new authors and organizations without prior connections to the community involving the generation of research clusters de novo and an overall diversification of the network. These findings suggest an ongoing evolution of the OSP community toward a more segmented, diverse, and inclusive network.

Accelerated Engineering of ELP-based Materials through Hybrid Biomimetic-De Novo Predictive Molecular Design.

Efforts to engineer high-performance protein-based materials inspired by nature have mostly focused on altering naturally occurring sequences to confer the desired functionalities, whereas de novo design lags significantly behind and calls for unconventional innovative approaches. Here, using partially disordered elastin-like polypeptides (ELPs) as initial building blocks we show that de novo engineering of protein materials can be accelerated through hybrid biomimetic design, which we achieve by integrating computational modeling, deep neural network, and recombinant DNA technology. This generalizable approach involves incorporating a series of de novo-designed sequences with α-helical conformation and genetically encoding them into biologically inspired intrinsically disordered repeating motifs. The new ELP variants maintained structural conformation and showed tunable supramolecular self-assembly out of thermal equilibrium with phase behavior in vitro. We illustrate the effective translation of the predicted molecular designs in structural and functional materials. The proposed methodology can be applied to a broad range of partially disordered biomacromolecules and potentially pave the way toward the discovery of novel structural proteins. This article is protected by copyright. All rights reserved.

Improvement of Perovskite Solar Cells Efficiency by Management of the Electron Withdrawing Groups in Hole Transport Materials: Theoretical Calculation and Experimental Verification.

Management of functional groups in hole transporting materials (HTMs) is a feasible strategy to improve perovskite solar cells (PSCs) efficiency. Therefore, starting from the carbazole-diphenylamine-based JY7 molecule, JY8 and JY9 molecules are incorporated into the different electron-withdrawing groups of fluorine and cyano groups on the side chains. The theoretical results reveal that the introduction of electron-withdrawing groups of JY8 and JY9 can improve these highest occupied molecular orbital(HOMO)energy levels, intermolecular stacking arrangements, and stronger interface adsorption on the perovskite. Especially, the results of molecular dynamics (MD) indicate that the fluorinated JY8 molecule can yield a preferred surface orientation, which exhibits stronger interface adsorption on the perovskite. To validate the computational model, the JY7-JY9 are synthesized and assembled into PSC devices. Experimental results confirm that the HTMs of JY8 exhibit outstanding performance, such as high hole mobility, low defect density, and efficient hole extraction. Consequently, the PSC devices based on JY8 achieve a higher PCE than those of JY7 and JY9. This work highlights the management of the electron-withdrawing groups in HTMs to realize the goal of designing HTMs for the improvement of PSC efficiency.

Tuning Two-Dimensional Phthalocyanine Dual Site Metal-Organic Framework Catalysts for the Oxygen Reduction Reaction.

Metal-organic frameworks (MOFs) offer an interesting opportunity for catalysis, particularly for metal-nitrogen-carbon (M-N-C) motifs by providing an organized porous structural pattern and well-defined active sites for the oxygen reduction reaction (ORR), a key need for hydrogen fuel cells and related sustainable energy technologies. In this work, we leverage electrochemical testing with computational models to study the electronic and structural properties in the MOF systems and their relationship to ORR activity and stability based on dual transitional metal centers. The MOFs consist of two M1 metals with amine nodes coordinated to a single M2 metal with a phthalocyanine linker, where M1/M2 = Co, Ni, or Cu. Co-based metal centers, in particular Ni-Co, demonstrate the highest overall activity of all nine tested MOFs. Computationally, we identify the dominance of Co sites, relative higher importance of the M2 site, and the role of layer M1 interactions on the ORR activity. Selectivity measurements indicate that M1 sites of MOFs, particularly Co, exhibit the lowest (<4%), and Ni demonstrates the highest (>46%) two-electron selectivity, in good agreement with computational studies. Direct in situ stability characterization, measuring dissolved metal ions, and calculations, using an alkaline stability metric, confirm that Co is the most stable metal in the MOF, while Cu exhibits notable instability at the M1. Overall, this study reveals how atomistic coupling of electronic and structural properties affects the ORR performance of dual site MOF catalysts and opens new avenues for the tunable design and future development of these systems for practical electrochemical applications.

A deep learning method for comparing Bayesian hierarchical models.

Bayesian model comparison (BMC) offers a principled approach to assessing the relative merits of competing computational models and propagating uncertainty into model selection decisions. However, BMC is often intractable for the popular class of hierarchical models due to their high-dimensional nested parameter structure. To address this intractability, we propose a deep learning method for performing BMC on any set of hierarchical models which can be instantiated as probabilistic programs. Since our method enables amortized inference, it allows efficient re-estimation of posterior model probabilities and fast performance validation prior to any real-data application. In a series of extensive validation studies, we benchmark the performance of our method against the state-of-the-art bridge sampling method and demonstrate excellent amortized inference across all BMC settings. We then showcase our method by comparing four hierarchical evidence accumulation models that have previously been deemed intractable for BMC due to partly implicit likelihoods. Additionally, we demonstrate how transfer learning can be leveraged to enhance training efficiency. We provide reproducible code for all analyses and an open-source implementation of our method. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Computational model of the spatiotemporal synergetic system dynamics of calcium, IP3 and dopamine in neuron cells

Computational model of the spatiotemporal synergetic system dynamics of calcium, IP3 and dopamine in neuron cells

Integrating Computational and Biological Hemodynamic Approaches to Improve Modeling of Atherosclerotic Arteries.

Atherosclerosis is the primary cause of cardiovascular disease, resulting in mortality, elevated healthcare costs, diminished productivity, and reduced quality of life for individuals and their communities. This is exacerbated by the limited understanding of its underlying causes and limitations in current therapeutic interventions, highlighting the need for sophisticated models of atherosclerosis. This review critically evaluates the computational and biological models of atherosclerosis, focusing on the study of hemodynamics in atherosclerotic coronary arteries. Computational models account for the geometrical complexities and hemodynamics of the blood vessels and stenoses, but they fail to capture the complex biological processes involved in atherosclerosis. Different in vitro and in vivo biological models can capture aspects of the biological complexity of healthy and stenosed vessels, but rarely mimic the human anatomy and physiological hemodynamics, and require significantly more time, cost, and resources. Therefore, emerging strategies are examined that integrate computational and biological models, and the potential of advances in imaging, biofabrication, and machine learning is explored in developing more effective models of atherosclerosis.

Why are triploid quaking aspen (Populus tremuloides) common?

Quaking aspen is a clonal tree species that has mixed ploidy, often with high relative abundance of both diploids and triploids but no haploids or tetraploids. Triploids typically have low fertility, leaving their occurrence apparently unlikely from an evolutionary perspective, unless they provide a "triploid bridge" to generating higher-fitness tetraploids-which are not observed in this species. This study focused on how triploidy can be maintained in quaking aspen. A computational model was used to simulate gamete production, sexual reproduction, asexual reproduction, parent survival, and offspring survival in a population. All parameters were assumed to be cytotype-dependent and environment-independent. Sampling methods were used to identify parameter combinations consistent with observed cytotype frequencies. Many processes and parameter values were sufficient to yield a moderate frequency of triploids, and very few were necessary. The most plausible route involved higher triploid survival at the parent or offspring stage and limited unreduced gamete production by either diploid or triploid parents. Triploid fertility was helpful but not necessary. The coexistence of diploids and triploids in quaking aspen is statistically likely and promoted by the existenceof commonly observed, long-lived triploid clones. However, other mechanisms not captured by the model related to environmental variation could also occur. Further empirical data or more complex but difficult-to-parameterize models are needed to gain further insight.

Network pharmacology speaking to ethnopharmacology: new data on an ancient remedy

Network pharmacology as a “green approach”, predicting metabolite behaviours chemically and biologically and guid¬ing biological experimental design, is a new strategy aiming to uncover the mechanism of action of natural products as drug candidates. It provides a powerful way to identify novel mechanisms of natural products with potential thera¬peutic effects. This approach has emerged as a powerful tool to overcome the limitations of traditional methods, such as the ability to predict the adverse effects of a drug and the likelihood of failure during clinical trials, by applying systems biology principles to the field of pharmacology. This method combines the multi-omics dataset, computer modeling, and chemical biology so as to reveal pharmaceutical actions and guide drug discovery. Therefore, computer-aided drug design combined with network pharmacology can be viewed as a novel in silico screening ap¬proach to drug discovery, by utilising chemoinformatics, bioinformatics, structure biology, and chemical biology. This strategy includes target-based virtual screening - molecular docking, ligand similarity-based virtual screening, and inverse screening (Inver-dock), providing a powerful tool for target identification of drug candidates, multitarget dis¬covery, and natural bioactive product profiling. It can also be used for selectivity profiling of drugs, drug repositioning, safety profiling, and metabolism profiling prediction (ADMET).

Challenges and directions for digital twin implementation in otorhinolaryngology.

Digital twin technology heralds a transformative era in Otorhinolaryngology (ORL), merging the physical and digital worlds to offer dynamic, virtual models of physical entities or processes. These models, capable of simulating, predicting, and optimizing real-world counterparts, are evolving from static replicas to intelligent, adaptive systems. Fueled by advancements in communication, sensor technology, big data analytics, Internet of Things (IoT), and simulation technologies, artificial intelligence (AI), digital twins in ORL promise personalized treatment planning, virtual experimentation, and therapeutic intervention optimization. Despite their potential, the integration of digital twins in ORL faces challenges including data privacy and security, data integration and interoperability, computational demands, model validation and accuracy, ethical and regulatory considerations, patient engagement, and cost and accessibility issues. Overcoming these challenges requires robust data protection measures, seamless data integration, substantial computational resources, rigorous validation studies, ethical transparency, patient education, and making the technology accessible and affordable. Looking ahead, the future of digital twins in ORL is bright, with advancements in AI and machine learning, omics data integration, real-time monitoring, virtual clinical trials, patient empowerment, seamless healthcare integration, longitudinal data analysis, and collaborative research. These developments promise to refine diagnostic and treatment strategies, enhance patient care, and facilitate more efficient and tailored ORL research, ultimately leading to more effective and personalized ORL management.

Microfoundations of the weakening educational gradient in fertility

The disappearance of the social gradient in fertility represents a paradigm shift that has called into question the validity of theories that predicted a decline in fertility with increased access to education and resources. Emerging theories have tried to explain this trend by highlighting a potential change in the fertility preferences of more educated couples. In this paper we add additional elements to this explanation. Using a computational modelling approach, we show that it is still possible to simulate the weakening social gradient in fertility, in the context of steady declines in family size preferences. Our results show that one of the key drivers of the change in the education–fertility relationship can be found in the transition to an increasingly regulated fertility regime. As the share of unplanned births decreases over time, the negative association between education and fertility weakens and the mechanisms that positively connect educational attainment with desired fertility become dominant.

Towards sharing tools and artefacts for reusable simulations in healthcare

ABSTRACT Discrete-event simulation (DES) is a widely used computational method in health services and health economic studies. Despite increasing recognition of the advantages of open, reusable DES models for both healthcare practitioners and simulation researchers, in practice very few authors share their model code alongside a published paper. In the context of Free and Open Source Software (FOSS), this paper presents a pilot framework called STARS: Sharing Tools and Artefacts for Reusable Simulations to begin to address the challenges and leverage the opportunities of sharing DES models in healthcare. STARS aligns with existing guidelines and documentation, including reproducibility initiatives, and enables computer models to be shared with users of differing technical abilities. We demonstrate the feasibility and applicability of STARS with three applied DES examples using Python. Our framework supports the development of open, reusable DES models which can enable partner healthcare organisations to preview, validate, and use models. Academic research teams can benefit from knowledge exchange, enhanced recognition and scrutiny of their work, and long-term archiving of models.

Personalized whole-brain neural mass models reveal combined Aβ and tau hyperexcitable influences in Alzheimer’s disease

Neuronal dysfunction and cognitive deterioration in Alzheimer’s disease (AD) are likely caused by multiple pathophysiological factors. However, mechanistic evidence in humans remains scarce, requiring improved non-invasive techniques and integrative models. We introduce personalized AD computational models built on whole-brain Wilson-Cowan oscillators and incorporating resting-state functional MRI, amyloid-β (Aβ) and tau-PET from 132 individuals in the AD spectrum to evaluate the direct impact of toxic protein deposition on neuronal activity. This subject-specific approach uncovers key patho-mechanistic interactions, including synergistic Aβ and tau effects on cognitive impairment and neuronal excitability increases with disease progression. The data-derived neuronal excitability values strongly predict clinically relevant AD plasma biomarker concentrations (p-tau217, p-tau231, p-tau181, GFAP) and grey matter atrophy obtained through voxel-based morphometry. Furthermore, reconstructed EEG proxy quantities show the hallmark AD electrophysiological alterations (theta band activity enhancement and alpha reductions) which occur with Aβ-positivity and after limbic tau involvement. Microglial activation influences on neuronal activity are less definitive, potentially due to neuroimaging limitations in mapping neuroprotective vs detrimental activation phenotypes. Mechanistic brain activity models can further clarify intricate neurodegenerative processes and accelerate preventive/treatment interventions.

Development of a Machine Learning (ML)-Based Computational Model to Estimate the Engineering Properties of Portland Cement Concrete (PCC)

AbstractPortland cement concrete (PCC) is the construction material most used worldwide. Hence, its proper characterization is fundamental for the daily-basis engineering practice. Nonetheless, the experimental measurements of the PCC’s engineering properties (i.e., Poisson’s Ratio -v-, Elastic Modulus -E-, Compressive Strength -ComS-, and Tensile Strength -TenS-) consume considerable amounts of time and financial resources. Therefore, the development of high-precision indirect methods is fundamental. Accordingly, this research proposes a computational model based on deep neural networks (DNNs) to simultaneously predict the v, E, ComS, and TenS. For this purpose, the Long-Term Pavement Performance database was employed as the data source. In this regard, the mix design parameters of the PCC are adopted as input variables. The performance of the DNN model was evaluated with 1:1 lines, goodness-of-fit parameters, Shapley additive explanations assessments, and running time analysis. The results demonstrated that the proposed DNN model exhibited an exactitude higher than 99.8%, with forecasting errors close to zero (0). Consequently, the machine learning-based computational model designed in this investigation is a helpful tool for estimating the PCC’s engineering properties when laboratory tests are not attainable. Thus, the main novelty of this study is creating a robust model to determine the v, E, ComS, and TenS by solely considering the mix design parameters. Likewise, the central contribution to the state-of-the-art achieved by the present research effort is the public launch of the developed computational tool through an open-access GitHub repository, which can be utilized by engineers, designers, agencies, and other stakeholders.