Coarse-grained Level Research Articles

A coarse-grained parcel method for heat and mass transfer simulations of spray coating processes

The Discrete Element Method (DEM) is commonly used for modeling the flow of particulate materials. Unfortunately, such detailed simulations are computationally very demanding, restricting its use for industrially-scaled processes. The number of particles in a simulation can be reduced by introducing parcels (i.e., “coarse graining”), which – in essence – relies on the increase of the particle diameter for interaction calculations. However, sophisticated models are necessary to preserve the original behavior of the material when using such an approach. Our present contribution extends available coarse-graining concepts by introducing models for (i) particle–fluid mass transfer and (ii) the deposition rate of spray droplets on particles. Our mass transfer model is based on an existing model for heat transfer. For the spray deposition model, we introduce an effective particle diameter to compute the correct amount of droplets that impact particles. We show that these models can be used with confidence up to a coarse-graining level of 5, which we demonstrate for a simple-shaped fluidized bed. The models proposed by us are critical for detailed simulations of spray coating processes since they enable precise particle-droplet-air interaction modeling at low computational cost.

State-space renormalization group theory of nonequilibrium reaction networks: Exact solutions for hypercubic lattices in arbitrary dimensions.

Nonequilibrium reaction networks (NRNs) underlie most biological functions. Despite their diverse dynamic properties, NRNs share the signature characteristics of persistent probability fluxes and continuous energy dissipation even in the steady state. Dynamics of NRNs can be described at different coarse-grained levels. Our previous work showed that the apparent energy dissipation rate at a coarse-grained level follows an inverse power-law dependence on the scale of coarse-graining. The scaling exponent is determined by the network structure and correlation of stationary probability fluxes. However, it remains unclear whether and how the (renormalized) flux correlation varies with coarse-graining. Following Kadanoff's real space renormalization group (RG) approach for critical phenomena, we address this question by developing a state-space renormalization group theory for NRNs, which leads to an iterative RG equationfor the flux correlation function. In square and hypercubic lattices, we solve the RG equationexactly and find two types of fixed point solutions. There is a family of nontrivial fixed points where the correlation exhibits power-law decay, characterized by a power exponent that can take any value within a continuous range. There is also a trivial fixed point where the correlation vanishes beyond the nearest neighbors. The power-law fixed point is stable if and only if the power exponent is less than the lattice dimension n. Consequently, the correlation function converges to the power-law fixed point only when the correlation in the fine-grained network decays slower than r^{-n} and to the trivial fixed point otherwise. If the flux correlation in the fine-grained network contains multiple stable solutions with different exponents, the RG iteration dynamics select the fixed point solution with the smallest exponent. The analytical results are supported by numerical simulations. We also discuss a possible connection between the RG flows of flux correlation with those of the Kosterlitz-Thouless transition.

Fine-Grained Ship Recognition from the Horizontal View Based on Domain Adaptation.

Ship recognition is a fundamental and essential step in maritime activities, and it can be widely used in maritime rescue, vessel management, and other applications. However, most studies conducted in this area use synthetic aperture radar (SAR) images and space-borne optical images, and those studies utilizing visible images are limited to the coarse-grained level. In this study, we constructed a fine-grained ship dataset with real images and simulation images that consisted of five categories of ships. To solve the problem of low accuracy in fine-grained ship classification with different angles in visible images, a network based on domain adaptation and a transformer was proposed. Concretely, style transfer was first used to reduce the gap between the simulation images and real images. Then, with the goal of utilizing the simulation images to execute classification tasks on the real images, a domain adaptation network based on local maximum mean discrepancy (LMMD) was used to align the different domain distributions. Furthermore, considering the innate attention mechanism of the transformer, a vision transformer (ViT) was chosen as the feature extraction module to extract the fine-grained features, and a fully connected layer was used as the classifier. Finally, the experimental results showed that our network had good performance on the fine-grained ship dataset with an overall accuracy rate of 96.0%, and the mean average precision (mAP) of detecting first and then classifying with our network was 87.5%, which also verified the feasibility of using images generated by computer simulation technology for auxiliary training.

Fine-grained, aspect-based sentiment analysis on economic and financial lexicon

Extracting sentiment from news text, social media and blogs has recently gained increasing interest in economics and finance. Despite many successful applications of sentiment analysis (SA) in these domains, the range of semantic techniques employed is still limited and predominantly focused on the detection of sentiment at a coarse-grained level. This paper proposes a novel methodology for Fine-Grained Aspect-based Sentiment (FiGAS) analysis. The aim is to identify the sentiment associated with specific topics of interest in each sentence of a document and to assign real-valued polarity scores between -1 and +1 to those topics. The proposed approach is unsupervised and customised to the economic and financial domains by using a specialised lexicon provided by us along with the FiGAS source code. Our lexicon-based SA approach relies on a detailed set of semantic polarity rules that allow understanding of the origin of sentiment – in the spirit of the recent trend on Interpretable AI. We provide an in-depth comparison of the performance of the FiGAS algorithm with other popular lexicon-based SA approaches in predicting a humanly annotated dataset in the economic and financial domains. Our results indicate that FiGAS statistically outperforms the other methods by providing a sentiment score that is closer to those of human annotators.

SIRNet: Fine-Grained Surgical Interaction Recognition

Surgical interaction recognition (SIR) plays a crucial role in navigation decision support for minimally invasive surgery (MIS) or robot-assisted MIS. Currently, the research in SIR is at a coarse-grained level and barely considers the surgical interaction dependencies unrelated to endoscopic images. This work proposes a fine-grained SIR method named SIRNet aiming at predicting surgical interaction triplets. In the proposed SIRNet, a multi-head self-attention mechanism learns the relations among surgical interaction triplets without defining them before the training process, while a multi-head cross-attention mechanism learns the relationships between the endoscopic images and each triplet. The bipartite matching loss, which considers the permutation and combination of instruments, verbs, and targets, is adopted to make appropriate learning and prediction for each component in the surgical interaction triplet. Moreover, a weight attention module is designed to weigh the importance of each predicted surgical interaction triplet and each component in triplet when predicting final valid surgical interaction triplets. The experimental results show the proposed method improves the performance of fine-grained SIR. In addition, experiments also present the effectiveness of each module. The code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/cynerelee/SIRNet</uri> <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/cynerelee/SIRNet</uri> .

Intrinsic viscuit probability distribution functions for transport coefficients of liquids and solids.

A reformulation of the Green-Kubo expressions for the transport coefficients of liquids in terms of a probability distribution function (PDF) of short trajectory contributions, which were named "viscuits," has been explored in a number of recent publications. The viscuit PDF, P, is asymmetric on the two sides of the distribution. It is shown here using equilibrium 3D and 2D molecular dynamics simulations that the viscuit PDF of a range of simple molecular single component and mixture liquid and solid systems can be expressed in terms of the same intrinsic PDF (P0), which is derived from P with the viscuit normalized by the standard deviation separately on each side of the distribution. P0 is symmetric between the two sides and can be represented for not very small viscuit values by the same gamma distribution formulated in terms of a single disposable parameter. P0 tends to an exponential in the large viscuit wings. Scattergrams of the viscuits and their associated single trajectory correlation functions are shown to distinguish effectively between liquids, solids, and glassy systems. The so-called viscuit square root method for obtaining the transport coefficients is shown to be a useful probe of small and statistically zero self-diffusion coefficients of molecules in the liquid and solid states, respectively. The results of this work suggest that the transport coefficients have a common underlying physical origin, reflecting at a coarse-grained level the traversal statistics of the system through its high-dimensioned potential energy landscape.

Effects of Slip-Spring Parameters and Rouse Bead Density on Polymer Dynamics in Multichain Slip-Spring Simulations.

The multichain slip-spring (MCSS) model is one of the coarse-grained models of polymers developed in the niche between bead-spring models and tube type descriptions. In this model, polymers are represented by Rouse chains connected by virtual springs that temporally connect the chains, hop along the chain, and are constructed and annihilated at the chain ends. Earlier studies have shown that MCSS simulations can nicely reproduce entangled and unentangled polymer dynamics. However, the model parameters have been chosen arbitrarily, and their effects have not been reported. In this study, for the first time, we systematically investigated the effects of model parameters: fugacity of virtual springs, its intensity, and the Rouse bead density. We validated the employed simulation code by confirming that the statistics of the system follow the theoretical setup. Namely, the virtual spring density is correctly controlled, and polymer chains exhibit ideal chain statistics irrespective of the chosen parameter values. For diffusion and linear viscoelasticity, simulation results obtained for different parameters can be superposed with each other by conversion factors for the bead number per chain and units of length, time, and modulus. These conversion factors follow scaling laws concerning the number of Rouse segments between two consecutive anchoring points of virtual springs along the polymer chain. Besides, diffusion and viscoelasticity excellently agree with literature data for the standard bead-spring simulation. These results imply that the coarse-graining level for the MCSS model can be arbitrarily chosen and controlled by model parameters.

Kinetic interplay between droplet maturation and coalescence modulates shape of aged protein condensates

Biomolecular condensates formed by the process of liquid–liquid phase separation (LLPS) play diverse roles inside cells, from spatiotemporal compartmentalisation to speeding up chemical reactions. Upon maturation, the liquid-like properties of condensates, which underpin their functions, are gradually lost, eventually giving rise to solid-like states with potential pathological implications. Enhancement of inter-protein interactions is one of the main mechanisms suggested to trigger the formation of solid-like condensates. To gain a molecular-level understanding of how the accumulation of stronger interactions among proteins inside condensates affect the kinetic and thermodynamic properties of biomolecular condensates, and their shapes over time, we develop a tailored coarse-grained model of proteins that transition from establishing weak to stronger inter-protein interactions inside condensates. Our simulations reveal that the fast accumulation of strongly binding proteins during the nucleation and growth stages of condensate formation results in aspherical solid-like condensates. In contrast, when strong inter-protein interactions appear only after the equilibrium condensate has been formed, or when they accumulate slowly over time with respect to the time needed for droplets to fuse and grow, spherical solid-like droplets emerge. By conducting atomistic potential-of-mean-force simulations of NUP-98 peptides—prone to forming inter-protein beta-sheets—we observe that formation of inter-peptide beta-sheets increases the strength of the interactions consistently with the loss of liquid-like condensate properties we observe at the coarse-grained level. Overall, our work aids in elucidating fundamental molecular, kinetic, and thermodynamic mechanisms linking the rate of change in protein interaction strength to condensate shape and maturation during ageing.

Critical active dynamics is captured by a colored-noise driven field theory

Active matter may sometimes behave almost indistinguishably from equilibrium matter. This is particularly evident for some particle-based models and active field-theories close to a critical point which falls in the Ising universality class. Here we show however that, even when critical, active particles strongly violate the equilibrium fluctuation-dissipation in the high-wave-vector and high-frequency regime. Conversely, at larger spatiotemporal scales the theorem is progressively restored and the critical dynamics is in effective equilibrium. We develop a field-theoretical description of this scenario employing a space-time correlated noise field finding that the theory qualitatively captures the numerical results already at the Gaussian level. Moreover a dynamic renormalization group analysis shows that the correlated noise does not change the equilibrium critical exponents. Our results demonstrate that a correlated noise field is a fundamental ingredient to describe critical active matter at the coarse-grained level.

Bipolar reasoning in feedback pathways

Instead of the conventional 0 and 1 values, bipolar reasoning uses −1, 0, +1 to describe double-sided judgements in which neutral elements are halfway between positive and negative evaluations (e.g., “uncertain” lies between “impossible” and “totally sure”). We discuss the state-of-the-art in bipolar logics and recall two medieval forerunners, i.e., William of Ockham and Nicholas of Autrecourt, who embodied a bipolar mode of thought that is eminently modern. Starting from the trivial observation that “once a wheat sheaf is sealed and tied up, the packed down straws display the same orientation”, we work up a new theory of the bipolar nature of networks, suggesting that orthodromic (i.e., feedforward, bottom-up) projections might be functionally coupled with antidromic (i.e., feedback, top-down) projections via the mathematical apparatus of presheaves/globular sets. When an entrained oscillation such as a neuronal spike propagates from A to B, changes in B might lead to changes in A, providing unexpected antidromic effects. Our account points towards the methodological feasibility of novel neural networks in which message feedback is guaranteed by backpropagation mechanisms endowed in the same feedforward circuits. Bottom-up/top-down transmission at various coarse-grained network levels provides fresh insights in far-flung scientific fields such as object persistence, memory reinforcement, visual recognition, Bayesian inferential circuits and multidimensional activity of the brain. Implying that axonal stimulation by external sources might backpropagate and modify neuronal electric oscillations, our theory also suggests testable previsions concerning the optimal location of transcranial magnetic stimulation's coils in patients affected by drug-resistant epilepsy.

The downside and upside of including backbone flexibility for predicting protein and antibody binding

In this study we will address the following issues: 1. Can we adapt our coarse-grained folding algorithm, Upside, to predict binding? 2. What kinds of interactions are important for binding compared to folding? 3. What level of coarse graining can be tolerated? 4. Under what conditions backbone flexibility can hinder or help the docking process? 5. Is predicting folding or binding harder? We achieved substantially greater predictive performance on protein-protein docking through new potential energy terms for binding and training of our force field on a docking benchmark set, and are comptetive with established methods.

Nonequilibrium Thermodynamics of Polymeric Liquids via Atomistic Simulation.

The challenge of calculating nonequilibrium entropy in polymeric liquids undergoing flow was addressed from the perspective of extending equilibrium thermodynamics to include internal variables that quantify the internal microstructure of chain-like macromolecules and then applying these principles to nonequilibrium conditions under the presumption of an evolution of quasie equilibrium states in which the requisite internal variables relax on different time scales. The nonequilibrium entropy can be determined at various levels of coarse-graining of the polymer chains by statistical expressions involving nonequilibrium distribution functions that depend on the type of flow and the flow strength. Using nonequilibrium molecular dynamics simulations of a linear, monodisperse, entangled C1000H2002 polyethylene melt, nonequilibrium entropy was calculated directly from the nonequilibrium distribution functions, as well as from their second moments, and also using the radial distribution function at various levels of coarse-graining of the constituent macromolecular chains. Surprisingly, all these different methods of calculating the nonequilibrium entropy provide consistent values under both planar Couette and planar elongational flows. Combining the nonequilibrium entropy with the internal energy allows determination of the Helmholtz free energy, which is used as a generating function of flow dynamics in nonequilibrium thermodynamic theory.

Retina organoids: Window into the biophysics of neuronal systems.

With a kind of magnetism, the human retina draws the eye of neuroscientist and physicist alike. It is attractive as a self-organizing system, which forms as a part of the central nervous system via biochemical and mechanical cues. The retina is also intriguing as an electro-optical device, converting photons into voltages to perform on-the-fly filtering before the signals are sent to our brain. Here, we consider how the advent of stem cell derived in vitro analogs of the retina, termed retina organoids, opens up an exploration of the interplay between optics, electrics, and mechanics in a complex neuronal network, all in a Petri dish. This review presents state-of-the-art retina organoid protocols by emphasizing links to the biochemical and mechanical signals of in vivo retinogenesis. Electrophysiological recording of active signal processing becomes possible as retina organoids generate light sensitive and synaptically connected photoreceptors. Experimental biophysical tools provide data to steer the development of mathematical models operating at different levels of coarse-graining. In concert, they provide a means to study how mechanical factors guide retina self-assembly. In turn, this understanding informs the engineering of mechanical signals required to tailor the growth of neuronal network morphology. Tackling the complex developmental and computational processes in the retina requires an interdisciplinary endeavor combining experiment and theory, physics, and biology. The reward is enticing: in the next few years, retina organoids could offer a glimpse inside the machinery of simultaneous cellular self-assembly and signal processing, all in an in vitro setting.

Modelling structural properties of cyanine dye nanotubes at coarse-grained level.

Self-assembly is a ubiquitous process spanning from biomolecular aggregates to nanomaterials. Even though the resulting aggregates can be studied through experimental techniques, the dynamic pathways of the process and the molecular details of the final structures are not necessarily easy to resolve. Consequently, rational design of self-assembling aggregates and their properties remains extremely challenging. At the same time, modelling the self-assembly with computational methods is not trivial, because its spatio-temporal scales are usually beyond the limits of all-atom based simulations. The use of coarse-grained (CG) models can alleviate this limitation, but usually suffers from the lack of optimised parameters for the molecular constituents. In this work, we describe the procedure of parametrizing a CG Martini model for a cyanine dye (C8S3) that self-assembles into hollow double-walled nanotubes. First, we optimised the model based on quantum mechanics calculations and all-atom reference simulations, in combination with available experimental data. Then, we conducted random self-assembly simulations, and the performance of our model was tested on preformed assemblies. Our simulations provide information on the time-dependent local arrangement of this cyanine dye, when aggregates are being formed. Furthermore, we provide guidelines for designing and optimising parameters for similar self-assembling nanomaterials.

A deep recommendation model of cross-grained sentiments of user reviews and ratings

The matrix factorization model based on user-item rating data has been widely studied and applied in recommender systems. However, data sparsity, the cold-start problem, and poor explainability have restricted its performance. Textual reviews usually contain rich information about items’ features and users’ sentiments and preferences, which can solve the problem of insufficient information from only user ratings. However, most recommendation algorithms that take sentiment analysis of review texts into account are either fine- or coarse-grained, but not both, leading to uncertain accuracy and comprehensiveness regarding user preference. This study proposes a deep learning recommendation model (i.e., DeepCGSR) that integrates textual review sentiments and the rating matrix. DeepCGSR uses the review sets of users and items as a corpus to perform cross-grained sentiment analysis by combining fine- and coarse-grained levels to extract sentiment feature vectors for users and items. Deep learning technology is used to map between the extracted feature vector and latent factor through the rating-based matrix factorization model and obtain deep, nonlinear features to predict the user's rating of an item. Iterative experiments on e-commerce datasets from Amazon show that DeepCGSR consistently outperforms the recommendation models LFM, SVD++, DeepCoNN, TOPICMF, and NARRE. Overall, comparing with other recommendation models, the DeepCGSR model demonstrated improved evaluation results by 14.113% over LFM, 13.786% over SVD++, 9.920% over TOPICMF, 5.122% over DeepCoNN, and 2.765% over NARRE. Meanwhile, the DeepCGSR has great potential in fixing the overfitting and cold-start problems. Built upon previous studies and findings, the DeepCGSR is the state of the art, moving the design and development of the recommendation algorithms forward with improved recommendation accuracy.

Theory and Practice of Coarse-Grained Molecular Dynamics of Biologically Important Systems.

Molecular dynamics with coarse-grained models is nowadays extensively used to simulate biomolecular systems at large time and size scales, compared to those accessible to all-atom molecular dynamics. In this review article, we describe the physical basis of coarse-grained molecular dynamics, the coarse-grained force fields, the equations of motion and the respective numerical integration algorithms, and selected practical applications of coarse-grained molecular dynamics. We demonstrate that the motion of coarse-grained sites is governed by the potential of mean force and the friction and stochastic forces, resulting from integrating out the secondary degrees of freedom. Consequently, Langevin dynamics is a natural means of describing the motion of a system at the coarse-grained level and the potential of mean force is the physical basis of the coarse-grained force fields. Moreover, the choice of coarse-grained variables and the fact that coarse-grained sites often do not have spherical symmetry implies a non-diagonal inertia tensor. We describe selected coarse-grained models used in molecular dynamics simulations, including the most popular MARTINI model developed by Marrink’s group and the UNICORN model of biological macromolecules developed in our laboratory. We conclude by discussing examples of the application of coarse-grained molecular dynamics to study biologically important processes.

Using a Coarse-Grained Modeling Framework to Identify Oligomeric Motifs with Tunable Secondary Structure.

Coarse-grained modeling can be used to explore general theories that are independent of specific chemical detail. In this paper, we present cg_openmm, a Python-based simulation framework for modeling coarse-grained hetero-oligomers and screening them for structural and thermodynamic characteristics of cooperative secondary structures. cg_openmm facilitates the building of coarse-grained topology and random starting configurations, setup of GPU-accelerated replica exchange molecular dynamics simulations with the OpenMM software package, and features a suite of postprocessing thermodynamic and structural analysis tools. In particular, native contact analysis, heat capacity calculations, and free energy of folding calculations are used to identify and characterize cooperative folding transitions and stable secondary structures. In this work, we demonstrate the capabilities of cg_openmm on a simple 1-1 Lennard-Jones coarse-grained model, in which each residue contains 1 backbone and 1 side-chain bead. By scanning both nonbonded and bonded force-field parameter spaces at the coarse-grained level, we identify and characterize sets of parameters which result in the formation of stable helices through cooperative folding transitions. Moreover, we show that the geometries and stabilities of these helices can be tuned by manipulating the force-field parameters.

Shared-view and specific-view information extraction for recommendation

In various recommender systems, ratings and reviews are the main information to show user preferences. However, recommendation models that only use ratings, such as collaborative filtering, are vulnerable to data sparsity. And models only using review information will also suffer from the sparsity of reviews. On one hand, most ratings and reviews are interrelated and complementary, reviews may explain why a user gives a high or low rating to an item. On the other hand, ratings and reviews are numerical and textual information, respectively, and they reflect the preference of the user from a coarse-grained level and a fine-grained level A user may comment positively about some aspects of an item, even he gives a very low score to this item. There are specific information among each of them because of their heterogeneity. Therefore, it is possible to learn more accurate representation of users and items by effectively integrating ratings and text reviews from different views, that is, shared-view and specific-view. In this paper, we propose a Shared-view and Specific-view Information extraction model for Recommendation (SSIR), which integrates the information from reviews and interaction matrix to predict ratings Our model has two key components, including shared-view information extraction and specific-view exploitation. From the perspective of shared-view, SSIR jointly minimizes the loss of confusion adversarial and rating prediction loss to extract the shared information from reviews and user–item interaction matrix. For the specific-view part, SSIR applies orthogonal constraints on shared-view and specific-view modules to extract the discriminative features from reviews and interaction data. We fuse the features extracted from these two views to predict the final ratings. In addition, we use auxiliary reviews to deal with the sparsity problem of reviews. Experimental results on eight datasets show the effectiveness and robustness of our method, which could adapt to the recommendation scenarios with fewer reviews and ratings.

Coarse-to-fine pseudo supervision guided meta-task optimization for few-shot object classification

Few-Shot Learning (FSL) is a challenging and practical learning pattern, aiming to solve a target task which has only a few labeled examples. Currently, the field of FSL has made great progress, but largely in the supervised setting, where a large auxiliary labeled dataset is required for offline training. However, the unsupervised FSL (UFSL) problem where the auxiliary dataset is fully unlabeled has been seldom investigated despite of its significant value. This paper focuses on the more general and challenging UFSL problem and presents a novel method named Coarse-to-Fine Pseudo Supervision-guided Meta-Learning (C2FPS-ML) for unsupervised few-shot object classification. It first obtains prior knowledge from an unlabeled auxiliary dataset during unsupervised meta-training, and then use the prior knowledge to assist the downstream few-shot classification task. Coarse-to-Fine Pseudo Supervisions in C2FPS-ML aim to optimize meta-task sampling process in unsupervised meta-training stage which is one of the dominant factors for improving the performance of meta-learning based FSL algorithms. Human can learn new concepts progressively or hierarchically following the coarse-to-fine manners. By simulating this human’s behaviour, we develop two versions of C2FPS-ML for two different scenarios: one is natural object dataset and another one is other kinds of dataset (e.g., handwritten character dataset). For natural object dataset scenario, we propose to exploit the potential hierarchical semantics of the unlabeled auxiliary dataset to build a tree-like structure of visual concepts. For another scenario, progressive pseudo supervision is obtained by forming clusters in different similarity aspects and is represented by a pyramid-like structure. The obtained structure is applied as the supervision to construct meta-tasks in meta-training stage, and prior knowledge from the unlabeled auxiliary dataset is learned from the coarse-grained level to the fine-grained level. The proposed method sets the new state of the art on the gold-standard miniImageNet and achieves remarkable results on Omniglot while simultaneously increases efficiency.

Time parametrizations in long-range interacting Bose-Einstein condensates

Time-reparametrization invariance in general relativistic space-time does not allow us to single out a time in quantum mechanics in a mechanical way of measurement. Motivated by this problem, we examine this gauge invariance in the ground state of the quasi-stationary coarse-grained state of a long-range interacting closed system of identical or identified, macroscopic, and spatiotemporally inhomogeneous Bose–Einstein condensates in the thermodynamic and Newtonian limits. As a result, we find that it is a theoretical counterexample of this gauge invariance, except for proper-time translational invariance, at a coarse-grained level.