ترغب بنشر مسار تعليمي؟ اضغط هنا

StochSoCs: High performance biocomputing simulations for large scale Systems Biology

98   0   0.0 ( 0 )
 نشر من قبل Elias Kouskoumvekakis
 تاريخ النشر 2017
والبحث باللغة English




اسأل ChatGPT حول البحث

The stochastic simulation of large-scale biochemical reaction networks is of great importance for systems biology since it enables the study of inherently stochastic biological mechanisms at the whole cell scale. Stochastic Simulation Algorithms (SSA) allow us to simulate the dynamic behavior of complex kinetic models, but their high computational cost makes them very slow for many realistic size problems. We present a pilot service, named WebStoch, developed in the context of our StochSoCs research project, allowing life scientists with no high-performance computing expertise to perform over the internet stochastic simulations of large-scale biological network models described in the SBML standard format. Biomodels submitted to the service are parsed automatically and then placed for parallel execution on distributed worker nodes. The workers are implemented using multi-core and many-core processors, or FPGA accelerators that can handle the simulation of thousands of stochastic repetitions of complex biomodels, with possibly thousands of reactions and interacting species. Using benchmark LCSE biomodels, whose workload can be scaled on demand, we demonstrate linear speedup and more than two orders of magnitude higher throughput than existing serial simulators.



قيم البحث

اقرأ أيضاً

Motivation: SBML is the most widespread language for the definition of biochemical models. Although dozens of SBML simulators are available, there is a general lack of support to the integration of SBML models within open-standard general-purpose sim ulation ecosystems. This hinders co-simulation and integration of SBML models within larger model networks, in order to, e.g. enable in silico clinical trials of drugs, pharmacological protocols, or engineering artefacts such as biomedical devices against Virtual Physiological Human models. Modelica is one of the most popular existing open-standard general-purpose simulation languages, supported by many simulators. Modelica models are especially suited for the definition of complex networks of heterogeneous models from virtually all application domains. Models written in Modelica (and in 100+ other languages) can be readily exported into black-box Functional Mock-Up Units (FMUs), and seamlessly co-simulated and integrated into larger model networks within open-standard language-independent simulation ecosystems. Results: In order to enable SBML model integration within heterogeneous model networks, we present SBML2Modelica, a software system translating SBML models into well-structured, user-intelligible, easily modifiable Modelica models. SBML2Modelica is SBML Level 3 Version 2-compliant and succeeds on 96.47% of the SBML Test Suite Core (with a few rare, intricate and easily avoidable combinations of constructs unsupported and cleanly signalled to the user). Our experimental campaign on 613 models from the BioModels database (with up to 5438 variables) shows that the major open-source (general-purpose) Modelica and FMU simulators achieve performance comparable to state-of-the-art specialized SBML simulators. Availability and implementation: https://bitbucket.org/mclab/sbml2modelica
In this paper we suggest that, under suitable conditions, supervised learning can provide the basis to formulate at the microscopic level quantitative questions on the phenotype structure of multicellular organisms. The problem of explaining the robu stness of the phenotype structure is rephrased as a real geometrical problem on a fixed domain. We further suggest a generalization of path integrals that reduces the problem of deciding whether a given molecular network can generate specific phenotypes to a numerical property of a robustness function with complex output, for which we give heuristic justification. Finally, we use our formalism to interpret a pointedly quantitative developmental biology problem on the allowed number of pairs of legs in centipedes.
139 - Evan J. Molinelli 2013
We present a new experimental-computational technology of inferring network models that predict the response of cells to perturbations and that may be useful in the design of combinatorial therapy against cancer. The experiments are systematic series of perturbations of cancer cell lines by targeted drugs, singly or in combination. The response to perturbation is measured in terms of levels of proteins and phospho-proteins and of cellular phenotype such as viability. Computational network models are derived de novo, i.e., without prior knowledge of signaling pathways, and are based on simple non-linear differential equations. The prohibitively large solution space of all possible network models is explored efficiently using a probabilistic algorithm, belief propagation, which is three orders of magnitude more efficient than Monte Carlo methods. Explicit executable models are derived for a set of perturbation experiments in Skmel-133 melanoma cell lines, which are resistant to the therapeutically important inhibition of Raf kinase. The resulting network models reproduce and extend known pathway biology. They can be applied to discover new molecular interactions and to predict the effect of novel drug perturbations, one of which is verified experimentally. The technology is suitable for application to larger systems in diverse areas of molecular biology.
For many stochastic models of interest in systems biology, such as those describing biochemical reaction networks, exact quantification of parameter uncertainty through statistical inference is intractable. Likelihood-free computational inference tec hniques enable parameter inference when the likelihood function for the model is intractable but the generation of many sample paths is feasible through stochastic simulation of the forward problem. The most common likelihood-free method in systems biology is approximate Bayesian computation that accepts parameters that result in low discrepancy between stochastic simulations and measured data. However, it can be difficult to assess how the accuracy of the resulting inferences are affected by the choice of acceptance threshold and discrepancy function. The pseudo-marginal approach is an alternative likelihood-free inference method that utilises a Monte Carlo estimate of the likelihood function. This approach has several advantages, particularly in the context of noisy, partially observed, time-course data typical in biochemical reaction network studies. Specifically, the pseudo-marginal approach facilitates exact inference and uncertainty quantification, and may be efficiently combined with particle filters for low variance, high-accuracy likelihood estimation. In this review, we provide a practical introduction to the pseudo-marginal approach using inference for biochemical reaction networks as a series of case studies. Implementations of key algorithms and examples are provided using the Julia programming language; a high performance, open source programming language for scientific computing.
Processes involving multi-input multi-step reaction cascades are used in developing novel biosensing, biocomputing, and decision making systems. In various applications different changes in responses of the constituent processing steps (reactions) in a cascade are desirable in order to allow control of the systems response. Here we consider conversion of convex response to sigmoid by intensity filtering, as well as threshold filtering, and we offer a general overview of this field of research. Specifically, we survey rate equation modelling that has been used for enzymatic reactions. This allows us to design modified biochemical processes as network components with responses desirable in applications.
التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا