Predicting physical properties of complex multi-scale systems is a common challenge and demands analysis of various temporal and spatial scales. However, physics alone is often not sufficient due to lack of knowledge on certain details of the system. With sufficient data, however, machine learning techniques may aid. If data are yet relatively cumbersome to obtain, hybrid methods may come to the rescue. We focus in this report on using various types of neural networks (NN) including NN\'s into which physics information is encoded (PeNN\'s) and also studied effects of NN\'s hyperparameters. We apply the networks to predict the viscosity of an emulsion as a function of shear rate. We show that using various network performance metrics as the mean squared error and the coefficient of determination ( R 2 ) that the PeNN\'s always perform better than the NN\'s, as also confirmed by a Friedman test with a p-value smaller than 0.0002. The PeNN\'s capture extrapolation and interpolation very well, contrary to the NN\'s. In addition, we have found that the NN\'s hyperparameters including network complexity and optimization methods do not have any effect on the above conclusions. We suggest that encoding NN\'s with any disciplinary system based information yields promise to better predict properties of complex systems than NN\'s alone, which will be in particular advantageous for small numbers of data. Such encoding would also be scalable, allowing different properties to be combined, without repetitive training of the NN\'s.

This paper presents a comprehensive investigation of mesoporous Silica utilizing a multi-scale modeling approach under periodic boundary conditions integrated with machine learning algorithms. The study begins with Molecular Dynamics (MD) simulations to extract Silica\'s elastic properties and thermal conductivity at the nano-scale, employing the Tersoff potential. Subsequently, the derived material characteristics are applied to a series of generated porous Representative Volume Elements (RVEs) at the microscale. This phase involves the exploration of porosity and void shape effects on Silica\'s thermal and mechanical properties, considering inhomogeneities\' distributions along the X-axis and random dispersion of pore cells within a three-dimensional space. Furthermore, the influence of pore shape is examined by defining open and closed-cell models, encompassing spherical and ellipsoidal voids with aspect ratios of 2 and 4. To predict the properties of porous Silica, a shallow Artificial Neural Network (ANN) is deployed, utilizing geometric parameters of the RVEs and porosity. Subsequently, it is revealed that Silica\'s thermal and mechanical behavior is linked to pore geometry, distribution, and porosity model. Finally, to classify the behavior of porous Silica into three categories, quasi-isotropic, orthotropic, and transversely-isotropic, three methodologies of decision tree approach, K-Nearest Neighbors (KNN) algorithm, and Support Vector Machines (SVMs) are employed. Among these, SVMs employing a quadratic kernel function demonstrate robust performance in categorizing the thermal and mechanical behavior of porous Silica.

Mathematical and computational models of biological systems are increasingly complex, typically comprised of hybrid multi-scale methods such as ordinary differential equations, partial differential equations, agent-based and rule-based models, etc. These mechanistic models concurrently simulate detail at resolutions of whole host, multi-organ, organ, tissue, cellular, molecular, and genomic dynamics. Lacking analytical and numerical methods, solving complex biological models requires iterative parameter sampling-based approaches to establish appropriate ranges of model parameters that capture corresponding experimental datasets. However, these models typically comprise large numbers of parameters and therefore large degrees of freedom. Thus, fitting these models to multiple experimental datasets over time and space presents significant challenges. In this work we undertake the task of reviewing, testing, and advancing calibration practices across models and dataset types to compare methodologies for model calibration. Evaluating the process of calibrating models includes weighing strengths and applicability of each approach as well as standardizing calibration methods. Our work compares the performance of our model agnostic Calibration Protocol (CaliPro) with approximate Bayesian computing (ABC) to highlight strengths, weaknesses, synergies, and differences among these methods. We also present next-generation updates to CaliPro. We explore several model implementations and suggest a decision tree for selecting calibration approaches to match dataset types and modeling constraints.

In this article, a microscopic constitutive model is established that includes friction, plastic, and spring elements and has clear physical meaning. The friction unit reflects the mutual friction between crack surfaces, the plastic unit reflects the development of concrete plasticity, and the fracture of the spring unit reflects the formation and expansion of interior cracks in concrete. In addition, the integration of the random field theory into this model uncovers the physical underpinnings of the relationship between concrete\'s nonlinearity and randomness. The multi-scale modeling of the concrete static damage constitutive model is then realized once the parameters of the random field are discovered using the macro test results. In order to apply the model\'s applicability in finite element programs, a subroutine was ultimately constructed. The experimental data and the anticipated values from the numerical simulation are in good agreement, supporting the model\'s realism.

We propose a general mathematical and computational approach to study cellular transport driven by a group of kinesin motors. It is a framework for multi-scale modeling that integrates kinetic models of single kinesin motors, including detachment and reattachment events, to study group behaviors of several motors. By formulating the problem as a semi-Markov process and applying a central limit theorem, asymptotic velocity and diffusivity can be readily calculated, which offers considerable computational advantage over Monte Carlo simulations in tasks such as parameter sensitivity analysis and model selection. We demonstrate the method with some examples. The importance of incorporating the intrinsic microscopic-level dynamics of individual motors is illustrated by showing how changes at the microscopic level propagate to the motor-cargo complex at a mesoscopic level. Particularly, we showcase an example in which changes in the second moment of single-motor characteristics gives rise to different first moment characteristics of the motor group.

Objective.Relative biological effectiveness (RBE) plays a vital role in carbon ion radiotherapy, which is a promising treatment method for reducing toxic effects on normal tissues and improving treatment efficacy. It is important to have an effective and precise way of obtaining RBE values to support clinical decisions. A method of calculating RBE from a mechanistic perspective is reported.Approach.Ratio of dose to obtain the same number of double strand breaks (DSBs) between different radiation types was used to evaluate RBE. Package gMicroMC was used to simulate DSB yields. The DSB inductions were then analyzed to calculate RBE. The RBE values were compared with experimental results.Main results.Furusawa\'s experiment yielded RBE values of 1.27, 2.22, 3.00 and 3.37 for carbon ion beam with dose-averaged LET of 30.3 keVμm-1, 54.5 keVμm-1, 88 keVμm-1and 137 keVμm-1, respectively. RBE values computed from gMicroMC simulations were 1.75, 2.22, 2.87 and 2.97. When it came to a more sophisticated carbon ion beam with 6 cm spread-out Bragg peak, RBE values were 1.61, 1.63, 2.19 and 2.36 for proximal, middle, distal and distal end part, respectively. Values simulated by gMicroMC were 1.50, 1.87, 2.19 and 2.34. The simulated results were in reasonable agreement with the experimental data.Significance.As a mechanistic way for the evaluation of RBE for carbon ion radiotherapy by combining the macroscopic simulation of energy spectrum and microscopic simulation of DNA damages, this work provides a promising tool for RBE calculation supporting clinical applications such as treatment planning.

The layered fibers of carbon-fiber-reinforced polymer (CFRP) composites exhibit low thermal conductivity (TC) throughout their thickness due to the poor TC of the polymeric resin. Improved heat transmission inside the hydrogen storage tank during the filling process can reduce further compression work, and improved heat insulation can minimize energy loss. Therefore, it is crucial to understand the thermal properties of composites. This paper reports the thermal behavior of plain-woven CFRP composite using simulation at the micro-, meso-, and macro-scales. The TC was predicted numerically and compared to experimental findings and analytical models. Good results were found. Using the approach of multi-scale modeling, a parametric study was carried out to analyze in depth the influence of certain variables on thermal properties. The study revealed that both fiber volume fraction and temperature significantly influenced the TC of the composite, with the interphase fiber/matrix thickness following closely in terms of impact. The matrix porosity was found to have a relatively slighter impact, particularly within the porosity range of 5 to 15%.

DNA supercoiling is central to many fundamental processes of living organisms. Its average level along the chromosome and over time reflects the dynamic equilibrium of opposite activities of topoisomerases, which are required to relax mechanical stresses that are inevitably produced during DNA replication and gene transcription. Supercoiling affects all scales of the spatio-temporal organization of bacterial DNA, from the base pair to the large scale chromosome conformation. Highlighted in vitro and in vivo in the 1960s and 1970s, respectively, the first physical models were proposed concomitantly in order to predict the deformation properties of the double helix. About fifteen years later, polymer physics models demonstrated on larger scales the plectonemic nature and the tree-like organization of supercoiled DNA. Since then, many works have tried to establish a better understanding of the multiple structuring and physiological properties of bacterial DNA in thermodynamic equilibrium and far from equilibrium. The purpose of this essay is to address upcoming challenges by thoroughly exploring the relevance, predictive capacity, and limitations of current physical models, with a specific focus on structural properties beyond the scale of the double helix. We discuss more particularly the problem of DNA conformations, the interplay between DNA supercoiling with gene transcription and DNA replication, its role on nucleoid formation and, finally, the problem of scaling up models. Our primary objective is to foster increased collaboration between physicists and biologists. To achieve this, we have reduced the respective jargon to a minimum and we provide some explanatory background material for the two communities.

Introduction: The lower esophageal sphincter (LES) controls the passage into the stomach and prevents reflex of contents into the esophagus. Dysfunctions of this region typically involves impairment of muscular function, leading to diseases including gastro-esophageal reflux disease and achalasia. The main objective of this study was to develop a finite element model from a unique human LES dataset reconstructed from an ultra-mill imaging setup, and then to investigate the effect of anatomical characteristics on intraluminal pressures. Methods: A pipeline was developed to generate a mesh from a set of input images, which were extracted from a unique ultra-mill sectioned human LES. A total of 216 nodal points with cubic Hermite basis function was allocated to reconstruct the LES, including the longitudinal and circumferential muscles. The resultant LES mesh was used in biomechanical simulations, utilizing a previously developed LES mathematical model based on the Visible Human data to calculate intraluminal pressures. Anatomical and functional comparisons were made between the Ultra-mill and Visible human models. Results: Overall, the Ultra-mill model contained lower cavity (1,796 vs. 5,400 mm3) and muscle (1,548 vs. 15,700 mm3) volumes than the Visible Human model. The Ultra-mill model also developed a higher basal pressure (13.8 vs. 14.7 mmHg) and magnitude of pressure (19.8 vs. 18.9 mmHg) during contraction. Out of all the geometric transformations (i.e., uniform enlargement of volume, lengthening along the center-axis, dilation of the diameter, and increasing muscle thickness), the muscle volume was found to be the main contributor of basal and magnitude of pressures. Increases in length also caused proportional increases to pressures, while dilation of diameter had a less influential reverse effect. Discussion: The findings provide information on interindividual variability in LES pressure and demonstrates that anatomy has a large influence on pressures. This model forms the basis of more complex simulations involving food bolus transport and predicting LES dysfunctions.

Fetal neuroinflammation and prenatal stress (PS) may contribute to lifelong neurological disabilities. Astrocytes and microglia, among the brain\'s non-neuronal \"glia\" cell populations, play a pivotal role in neurodevelopment and predisposition to and initiation of disease throughout lifespan. One of the most common neurodevelopmental disorders manifesting between 1-4 years of age is the autism spectrum disorder (ASD). A pathological glial-neuronal interplay is thought to increase the risk for clinical manifestation of ASD in at-risk children, but the mechanisms remain poorly understood, and integrative, multi-scale models are needed. We propose a model that integrates the data across the scales of physiological organization, from genome to phenotype, and provides a foundation to explain the disparate findings on the genomic level. We hypothesize that via gene-environment interactions, fetal neuroinflammation and PS may reprogram glial immunometabolic phenotypes that impact neurodevelopment and neurobehavior. Drawing on genomic data from the recently published series of ovine and rodent glial transcriptome analyses with fetuses exposed to neuroinflammation or PS, we conducted an analysis on the Simons Foundation Autism Research Initiative (SFARI) Gene database. We confirmed 21 gene hits. Using unsupervised statistical network analysis, we then identified six clusters of probable protein-protein interactions mapping onto the immunometabolic and stress response networks and epigenetic memory. These findings support our hypothesis. We discuss the implications for ASD etiology, early detection, and novel therapeutic approaches. We conclude with delineation of the next steps to verify our model on the individual gene level in an assumption-free manner. The proposed model is of interest for the multidisciplinary community of stakeholders engaged in ASD research, the development of novel pharmacological and non-pharmacological treatments, early prevention, and detection as well as for policy makers.