Browsing by Author "Svičević, Marina"

Now showing 1 - 10 of 10

Coupling finite element and huxley models in multiscale muscle modeling
(2015) Stojanović, Boban; Svičević, Marina; Kaplarević-Mališić, Ana; Ivanović, Miloš; Nedic D.; Filipovic, Nenad; Mijailovich S.
© 2015 IEEE. In this paper we present a novel approach in multi-scale muscle modeling based on finite element method and Huxley crossbridge kinetics model. In order to determine the mechanical response of a muscle, we implement basic mechanical principles of motion of deformable bodies using finite element method. Constitutive properties of muscle are defined by the number of molecular interconnections between the myosin and actin filaments. To account for these effects, we used Huxley's micro model based on sliding filament theory to calculate muscle active forces and instantaneous stiffnesses in FE integration points. In order to run these computationally expensive simulations we have also developed a special parallelization strategy which gives speedup of two orders of magnitude. Results obtained using presented multi-scale model are compared to those obtained by Hill's phenomenological model.
Dilated cardiomyopathy myosin mutants have reduced force-generating capacity
(2018) Ujfalusi Z.; Vera C.; Mijailovich S.; Svičević, Marina; Yu E.; Kawana M.; RUPPEL K.; Spudich J.; Geeves, Michael; LEINWAND L.
© 2018 Ujfalusi et al. Dilated cardiomyopathy (DCM) and hypertrophic cardiomyopathy (HCM) can cause arrhythmias, heart failure, and cardiac death. Here, we functionally characterized the motor domains of five DCM-causing mutations in human -cardiac myosin. Kinetic analyses of the individual events in the ATPase cycle revealed that each mutation alters different steps in this cycle. For example, different mutations gave enhanced or reduced rate constants of ATP binding, ATP hydrolysis, or ADP release or exhibited altered ATP, ADP, or actin affinity. Local effects dominated, no common pattern accounted for the similar mutant phenotype, and there was no distinct set of changes that distinguished DCM mutations from previously analyzed HCM myosin mutations. That said, using our data to model the complete ATPase contraction cycle revealed additional critical insights. Four of the DCM mutations lowered the duty ratio (the ATPase cycle portion when myosin strongly binds actin) because of reduced occupancy of the force-holding AMD complex in the steady state. Under load, the AMD state is predicted to increase owing to a reduced rate constant for ADP release, and this effect was blunted for all five DCM mutations. We observed the opposite effects for two HCM mutations, namely R403Q and R453C. Moreover, the analysis predicted more economical use of ATP by the DCM mutants than by WT and the HCM mutants. Our findings indicate that DCM mutants have a deficit in force generation and force-holding capacity due to the reduced occupancy of the force-holding state.
Employing phenomenological model in load-balancing optimization of parallel multi-scale muscle simulations
(2015) Kaplarević-Mališić, Ana; Ivanović, Miloš; Stojanović, Boban; Svičević, Marina; Antonijević Đ.
© 2015 IEEE. Since multi-scale models of muscles rely on the integration of physical and biochemical properties across multiple length and time scales, these models are highly CPU consuming and memory intensive. Therefore, their practical implementation and usage in real-world applications is limited by their high requirements for computational power. There are various reported solutions to the problems of the distributed computation of the complex systems that could also be applied to the multi-scale muscle simulations. In this paper, we present a novel load balancing method for parallel multi-scale muscle simulations on distributed computing resources. The method uses data obtained from simple Hill phenomenological model in order to predict computational weights of the integration points within the multi-scale model. Using obtained weights it is possible to improve domain decomposition prior to multi-scale simulation run and consequently significantly reduce computational time. The method is applied to two-scale muscle model where a finite element (FE) macro model is coupled with Huxley's model of cross-bridge kinetics on the microscopic level. The massive parallel solution is based on decomposition of micro model domain and static scheduling policy. It was verified on real-world example, showing high utilization of all involved CPUs and ensuring high scalability, thanks to the novel scheduling approach. Performance analysis clearly shown that inclusion of complexities prediction in reducing the execution time of parallel run by about 40% compared to the same model with scheduler that assumes equal complexities of all micro models.
Machine learned domain decomposition scheme applied to parallel multi-scale muscle simulation
(2019) Ivanović, Miloš; Kaplarević-Mališić, Ana; Stojanović, Boban; Svičević, Marina; Mijailovich, Srboljub
Since multi-scale models of muscles rely on the integration of physical and biochemical properties across multiple length and time scales, they are highly processor and memory intensive. Consequently, their practical implementation and usage in real-world applications is limited by high computational requirements. There are various reported solutions to the problem of parallel computation of various multi-scale models, but due to their inherent complexity, load balancing remains a challenging task. In this article, we present a novel load balancing method for multi-scale simulations based on finite element (FE) method. The method employs a computationally simple single-scale model and machine learning in order to predict computational weights of the integration points within a complex multi-scale model. Employing the obtained weights, it is possible to improve the domain decomposition prior to the complex multi-scale simulation run and consequently reduce computation time. The method is applied to a two-scale muscle model, where the FE on macroscale is coupled with Huxley’s model of cross-bridge kinetics on the microscale. Our massive parallel solution is based on the static domain decomposition policy and operates in a heterogeneous (central processing units + graphics processing units) environment. The approach has been verified on a real-world example of the human tongue, showing high utilization of all processors and ensuring high scalability, owing to the proposed load balancing scheme. The performance analysis shows that the inclusion of the prediction of the computational weights reduces execution time by about 40% compared to the run which uses a trivial load balancer which assumes identical computational weights of all micro-models. The proposed domain decomposition approach possesses a high capability to be applied in a variety of multi-scale models based on the FE method.
Modeling the Actin.myosin ATPase Cross-Bridge Cycle for Skeletal and Cardiac Muscle Myosin Isoforms
(2017) Mijailovich S.; Nedic D.; Svičević, Marina; Stojanović, Boban; Walklate J.; Ujfalusi Z.; Geeves, Michael
© 2017 Biophysical Society Modeling the complete actin.myosin ATPase cycle has always been limited by the lack of experimental data concerning key steps of the cycle, because these steps can only be defined at very low ionic strength. Here, using human β-cardiac myosin-S1, we combine published data from transient and steady-state kinetics to model a minimal eight-state ATPase cycle. The model illustrates the occupancy of each intermediate around the cycle and how the occupancy is altered by changes in actin concentration for [actin] = 1–20Km. The cycle can be used to predict the maximal velocity of contraction (by motility assay or sarcomeric shortening) at different actin concentrations (which is consistent with experimental velocity data) and predict the effect of a 5 pN load on a single motor. The same exercise was repeated for human α-cardiac myosin S1 and rabbit fast skeletal muscle S1. The data illustrates how the motor domain properties can alter the ATPase cycle and hence the occupancy of the key states in the cycle. These in turn alter the predicted mechanical response of the myosin independent of other factors present in a sarcomere, such as filament stiffness and regulatory proteins. We also explore the potential of this modeling approach for the study of mutations in human β-cardiac myosin using the hypertrophic myopathy mutation R453C. Our modeling, using the transient kinetic data, predicts mechanical properties of the motor that are compatible with the single-molecule study. The modeling approach may therefore be of wide use for predicting the properties of myosin mutations.
Multi-modeling and multi-scale modeling as tools for solving complex realworld problems
(2016) Stojanović, Boban; Ivanović, Miloš; Kaplarević-Mališić, Ana; Simic, Visnja; Milivojevic̀ M.; Nedic D.; Svičević, Marina; Milivojevic N.; Mijailovich S.
In previous decades a number of computational methods for calculation of very complex physical phenomena with a satisfactory accuracy have been developed. Most of these methods usually model only a single physical phenomenon, while their performance regarding accuracy and efficiency are limited within narrow spatial and temporal domains. However, solving realworld problems often requires simultaneous analysis of several coupled physical phenomena that extend over few spatial and temporal scales. Thus, in the last decade, simultaneous modeling a number of physical phenomena (multi-modeling) and modeling across few scales (multi-scale modeling) have gained a huge importance. In this paper we give an overview of multi-modeling and multi-scale methods developed during the last decade within the Group for Scientific Computing at Faculty of Science, University of Kragujevac. In addition, we give a short review of accompanying problems that we had to solve in order to make the methods applicable in practice, such as parallelization of computations, parameters calibration, etc. In the first part of the paper we present methods for modeling various aspects of muscle behavior and their coupling into complex multi-models. The mechanical behavior of muscles is derived from the behavior of many individual components working together across spatial and temporal scales. Capturing the interplay between these components resulted in efficient multiscale model. The rest of the paper is reserved for the presentation of multi-models for solving real-world problems in the field of water resources management, as well as methods for calibration of complex models parameters. As most illustrative example, we present methodology for solving the problem of water leakage under Visegrad dam at Drina River in Republic of Srpska. With the aim to support decision making process during dam remediation, we have developed specialized multi-model that continuously uses acquired observations to estimate spatial distribution of main karst conductors, their characteristics, as well as hydraulic variables of the system.
Multi-scale striated muscle contraction model linking sarcomere length-dependent cross-bridge kinetics to macroscopic deformation
(2020) Stojanović, Boban; Svičević, Marina; Kaplarević-Mališić, Ana; Gilbert, Richard; Mijailovich S.
© 2019 The investigation of healthy and diseased muscle behavior via in silico analysis requires the modeling of biophysical processes on multiple spatial and temporal scales. Owing to the complexity of the phenomena in question, simultaneous simulations of all the processes across different scales are extremely computationally expensive. Therefore, many multi-scale models utilize simplified phenomenological models at the micro level. However, such models may not be able to predict transient contractile behavior accurately when the deformation is unsteady or non-uniform. To overcome these deficiencies of phenomenological models, we propose a novel multi-scale muscle model in which continuum muscle mechanics are modeled utilizing the finite element method, and the material characteristics of muscle tissues at the microscopic scale are defined by Huxley's model of muscle contraction. Owing to the specific application of the sliding-filament theory coupled with the kinetic formulation of Gordon's length-tension relationship, the proposed model can provide more precise simulations of muscle behavior under both isotonic and transient conditions. The proposed model is verified using both benchmark data and real-world examples, and the results are compared to corresponding predictions obtained using the FE-Hill model. Specific implementations of biophysical components at the muscle fiber scale are validated by comparing them to predictions obtained using a spatially explicit molecular model implemented on the MUSICO platform. To enable the execution of two-scale simulations in a reasonable timeframe, we utilize a custom-tailored parallelization platform called Mexie. The ability of the proposed model to describe tissue-scale motor system behavior and the efficiency of its parallel execution are demonstrated through simulations of tongue movement during the propulsive phase of human swallowing. In these simulations the tissue's complex muscular structure is represented by a 2D finite element mesh. The proposed model provides tools for the scientific investigation of musculoskeletal disorders and facilitates the prospective development of clinical applications for characterizing neuromuscular disorders and monitoring disease progression during therapy.
Nebulin and titin modulate cross-bridge cycling and length-dependent calcium sensitivity
(2019) Mijailovich S.; Stojanović, Boban; Nedic D.; Svičević, Marina; Geeves, Michael; Irving T.; Granzier H.
© 2019 Mijailovich et al. Various mutations in the structural proteins nebulin and titin that are present in human disease are known to affect the contractility of striated muscle. Loss of nebulin is associated with reduced actin filament length and impairment of myosin binding to actin, whereas titin is thought to regulate muscle passive elasticity and is likely involved in length-dependent activation. Here, we sought to assess the modulation of muscle function by these sarcomeric proteins by using the computational platform muscle simulation code (MUSICO) to quantitatively separate the effects of structural changes, kinetics of cross-bridge cycling, and calcium sensitivity of the thin filaments. The simulations show that variation in thin filament length cannot by itself account for experimental observations of the contractility in nebulin-deficient muscle, but instead must be accompanied by a decreased myosin binding rate. Additionally, to match the observed calcium sensitivity, the rate of TnI detachment from actin needed to be increased. Simulations for cardiac muscle provided quantitative estimates of the effects of different titin-based passive elasticities on muscle force and activation in response to changes in sarcomere length and interfilament lattice spacing. Predicted force-pCa relations showed a decrease in both active tension and sensitivity to calcium with a decrease in passive tension and sarcomere length. We conclude that this behavior is caused by partial redistribution of the muscle load between active muscle force and titin-dependent passive force, and also by redistribution of stretch along the thin filament, which together modulate the release of TnI from actin. These data help advance understanding of how nebulin and titin mutations affect muscle function.
Numerical solution of Stefan problem with variable space grid method based on mixed finite element/finite difference approach
(2017) Ivanović, Miloš; Svičević, Marina; Savovic, Svetislav
The purpose of this paper is to improve the accuracy and stability of the existing solutions to 1D Stefan problem with time-dependent Dirichlet boundary conditions. The accuracy improvement should come with respect to both temperature distribution and moving boundary location. The variable space grid method based on mixed finite element/finite difference approach is applied on 1D Stefan problem with time-dependent Dirichlet boundary conditions describing melting process. The authors obtain the position of the moving boundary between two phases using finite differences, whereas finite element method is used to determine temperature distribution. In each time step, the positions of finite element nodes are updated according to the moving boundary, whereas the authors map the nodal temperatures with respect to the new mesh using interpolation techniques. The authors found that computational results obtained by proposed approach exhibit good agreement with the exact solution. Moreover, the results for temperature distribution, moving boundary location and moving boundary speed are more accurate th an those obtained by variable space grid method based on pure finite differences. The authors’ approach clearly differs from the previous solutions in terms of methodology. While pure finite difference variable space grid method produces stable solution, the mixed finite element/finite difference variable space grid scheme is significantly more accurate, especially in case of high alpha. Slightly modified scheme has a potential to be applied to 2D and 3D Stefan problems.
The ATPase cycle of human muscle myosin II isoforms: Adaptation of a single mechanochemical cycle for different physiological roles
(2019) Johnson C.; Walklate J.; Svičević, Marina; Mijailovich S.; Vera C.; Karabina A.; LEINWAND L.; Geeves, Michael
© 2019 Johnson et al. Striated muscle myosins are encoded by a large gene family in all mammals, including humans. These isoforms define several of the key characteristics of the different striated muscle fiber types, including maximum shortening velocity. We have previously used recombinant isoforms of the motor domains of seven different human myosin isoforms to define the actinmyosin cross-bridge cycle in solution. Here, we present data on an eighth isoform, the perinatal, which has not previously been characterized. The perinatal is distinct from the embryonic isoform, appearing to have features in common with the adult fast-muscle isoforms, including weak affinity of ADP for actinmyosin and fastADPrelease.Wego on to use a recently developed modeling approach, MUSICO, to explore how well the experimentally defined cross-bridge cycles for each isoform in solution can predict the characteristics of muscle fiber contraction, including duty ratio, shortening velocity, ATP economy, and load dependence of these parameters. The work shows that the parameters of the cross-bridge cycle predict many of the major characteristics of each muscle fiber type and raises the question of what sequence changes are responsible for these characteristics.