We've updated our Privacy Policy to make it clearer how we use your personal data.
We use cookies to provide you with a better experience, read our Cookie Policy
EP39611
Abstract: Background: Virtual patients (VP) simulations and analysis is a rapidly developing area of research, which plays a crucial role in improving clinical reasoning. Generating cohorts of virtual patients and running multiple simulations on large QSP models are computationally demanding and time consuming procedures. Hybrid hardware environment together with software, which supports parallel simulations setup, can drastically improve applicability of VP-based approach.
Objectives: One of the main challenges of VP simulations in the QSP field is solving large differential equation systems with thousands of input parameters sets multiplied by a number of therapies. The study proposes a hardware and software environment, which can efficiently address such problems.
Methods: VP simulations can be parallelized by distributing individual simulations on available computation nodes. HTCondor [1] is an open source cluster software used in the study to activate computation nodes and execute individual VP simulations. The software supports different operating systems and can optimally utilize resources available on-site. The ability of such on-site cluster computations is important from both cost saving prospective and security reasons of QSP projects. The on-site environment can be extended by adding cloud-based resources. Together they constitute an on-demand hybrid cloud infrastructure. The software part of environment utilizes the ability of Julia language to distribute workloads between worker nodes and parallel simulations features provided by SciML packages [2]. The key software component of the environment is an open-source package HetaSimulator [3], which provides an interface to set up and run parallel multi-conditional VP simulations.
Results: The proposed hardware and software environment sufficiently reduces VP simulation time by supporting parallel hybrid cloud setup. It allows the modelers to utilize computers available on-site and add cloud-based resources on demand. It also supports multi-conditional simulations to model different therapies.
Conclusions: Using HetaSimulator in a hybrid cloud environment can efficiently utilize both on-site and cloud resources to run parallel VP simulations with therapy-based conditions.
Summary: Using HetaSimulator in a hybrid cloud environment can efficiently utilize both on-site and cloud resources to run parallel VP simulations with therapy-based conditions.References: Thain D, Tannenbaum T, and Livny M. Distributed computing in practice: the condor experience. Concurr. Comput., vol. 17, no. 2-4, pp. 323–356, Feb. 2005.
Rackauckas C, Nie Q. DifferentialEquations.jl – A Performant and Feature-Rich Ecosystem for Solving Differential Equations in Julia. Journal of Open Research Software, 5(1):15, 2017.
Borisov I, Metelkin E. HetaProject – a Single Open Source Platform for Modeling, Simulation and Parameters Estimation in QSP. 2021. 10.13140/RG.2.2.31275.77608.
Objectives: One of the main challenges of VP simulations in the QSP field is solving large differential equation systems with thousands of input parameters sets multiplied by a number of therapies. The study proposes a hardware and software environment, which can efficiently address such problems.
Methods: VP simulations can be parallelized by distributing individual simulations on available computation nodes. HTCondor [1] is an open source cluster software used in the study to activate computation nodes and execute individual VP simulations. The software supports different operating systems and can optimally utilize resources available on-site. The ability of such on-site cluster computations is important from both cost saving prospective and security reasons of QSP projects. The on-site environment can be extended by adding cloud-based resources. Together they constitute an on-demand hybrid cloud infrastructure. The software part of environment utilizes the ability of Julia language to distribute workloads between worker nodes and parallel simulations features provided by SciML packages [2]. The key software component of the environment is an open-source package HetaSimulator [3], which provides an interface to set up and run parallel multi-conditional VP simulations.
Results: The proposed hardware and software environment sufficiently reduces VP simulation time by supporting parallel hybrid cloud setup. It allows the modelers to utilize computers available on-site and add cloud-based resources on demand. It also supports multi-conditional simulations to model different therapies.
Conclusions: Using HetaSimulator in a hybrid cloud environment can efficiently utilize both on-site and cloud resources to run parallel VP simulations with therapy-based conditions.
Summary: Using HetaSimulator in a hybrid cloud environment can efficiently utilize both on-site and cloud resources to run parallel VP simulations with therapy-based conditions.References: Thain D, Tannenbaum T, and Livny M. Distributed computing in practice: the condor experience. Concurr. Comput., vol. 17, no. 2-4, pp. 323–356, Feb. 2005.
Rackauckas C, Nie Q. DifferentialEquations.jl – A Performant and Feature-Rich Ecosystem for Solving Differential Equations in Julia. Journal of Open Research Software, 5(1):15, 2017.
Borisov I, Metelkin E. HetaProject – a Single Open Source Platform for Modeling, Simulation and Parameters Estimation in QSP. 2021. 10.13140/RG.2.2.31275.77608.
Ask the author a question about this poster.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Related Posters
5 Ways: Treatment of CAD
Suh Kyung (Chloe) Yu
The QSP model of microglia role in tau pathology and neurodegeneration
Stepan Lerner, Maria Myshkina, Tatiana Karelina
DNA-Stable Isotope Probing Technology (DNA-SIP)
Alex Brown
Suppository Bases
Helen
Physico-Chemical Characterization of Lyophilizates
Helen
