BioNessie

Category Cross-Omics>Pathway Analysis/Gene Regulatory Networks/Tools and Cross-Omics>Agent-Based Modeling/Simulation/Tools

Abstract BioNessie (BIOchemical NEtworks Simulation and AnalySIs Environment) is a platform-independent environment for modeling biochemical networks, and simulating and analyzing the dynamic behavior of biochemical models.

BioNessie uses a modular architecture that allows the developer to easily plug-in various components and update them. It was developed using Java technology and can run on any platform that supports JRE (Java Runtime Environment 1.5.x or higher).

BioNessie provides a user-friendly Graphical User Interface (GUI) which allows the user to import, create, edit and export biochemical models conforming to the Systems Biology Markup Language (SBML) standard.

LibSBML is being utilized to read, write, manipulate, translate, and validate SBML files and data streams.

LibSBML - LibSBML is a free, open-source programming library to help you read, write, manipulate, translate, and validate SBML files and data streams.

It is Not an application itself (though it does come with example programs), but rather a library you can embed in your own applications.

BioNessie’s unique Concurrent Versions System (CVS) design helps users keep track of the version history of their SBML models during construction and subsequent modification.

The core of BioNessie is comprised of the SBML ODE Solver Library (SOSlib). SOSlib provides a programming library for symbolic and numerical analysis of a system of ordinary differential equations (ODEs) derived from a chemical reaction network encoded in the SBML format.

SOSlib employs LibSBML’s Abstract Syntax Tree (AST) for formula representation to construct ODE systems, their Jacobian matrix and other derivatives.

SUNDIALS’ version of CVODE is incorporated for numerical integration and sensitivity analysis of stiff and non-stiff ODE systems.

SUNDIALS - SUNDIALS (SUite of Nonlinear and DIfferential/ALgebraic equation Solvers) is a suite of advanced computational codes for solving large-scale problems that can be modeled as a system of nonlinear algebraic equations, or as initial-value problems in ordinary differential or differential-algebraic equations.

The basic versions of these codes are called KINSOL, CVODE, and IDA, respectively. Also available are extensions of these codes for sensitivity analysis (CVODES and IDAS) and a MATLAB toolbox (sundialsTB) providing interfaces to CVODES, IDAS, and KINSOL.

BioNessie can generate the changes of ‘species amounts’ and parameter values over time by simulating the SBML model numerically with SOSlib.

The simulation results can be generated in two (2) ways: plots and report text files.

BioNessie is Not only an editor and simulator --

BioNessie is Not only an editor and simulator, but also an analyzer. It supports parameter scans, sensitivity analysis, and parameter estimation for biochemical networks.

These methods are commonly used approaches for understanding the dynamics of a biochemical model.

Parameter scanning - Parameter scanning permits the exploration of a model’s behavior over different ranges of parameter values. This is achieved by running simulations for specific values in the range.

Sensitivity analysis - Sensitivity analysis is the study of the response of system variables to changes in parameter values, which can be used to establish the contribution of individual parameter values to the overall performance of a complex biochemical system.

Parameter estimation - Parameter estimation is a method used to perform an estimation of biological parameters based on any observable wet-lab data.

The manufacturer uses a Genetic Algorithm (GA) to estimate the parameters by searching different rate constant sets in a predefined range to minimize the difference between the time-course data (obtained from wet lab) and simulation results of the model.

BioNessieG - A Grid Enabled version of BioNessie --

BioNessieG provides users with the scalability and availability needed for the simulation of large scale biochemical models incorporating parameter estimations and parameter scans leveraging the computational power of the Grid.

Parameter scanning and parameter estimation are quite time-consuming activities. In the parameter scanning process, BioNessie has to perform simulations many times. However, all the simulations can be run in parallel allowing for considerable speed up.

The same situation also applies to the parameter estimation. Due to their trivially parallel nature, large scale parameter scans and parameter estimation are well suited to execution on High-Performance Computing (HPC) cluster resources such as the National Grid Service (NGS), ScotGrid and Condor pools.

Note: Grid computing provides an infrastructure in which it is possible to share and aggregate computing resources from many diverse sources and view them as one virtual resource. These resources can be CPU, memory and storage amongst others.

BioNessieG Architecture - An Overview --

BioNessie has a modular architecture and extensions can be easily implemented by plugging a module into the BioNessie application.

The client side of BioNessieG has been designed as a graphical user interface for specifying the model and parameters while the Grid side is the computational engine which is fully Grid enabled. The client side has cross-platform compatibility because Java has been chosen as the software development language.

For the Grid side, the computational engine has been implemented on different Grid resources including the NGS, ScotGrid and Condor pools.

The system has been designed to be extensible so that other resources can be integrated seamlessly.

Each of these Grid resource head-nodes have been configured with SOSlib, LibSBML and SUNDIALS (see above…), as well as to accept inputs encoded in SBML formats.

The communication between the Grid resources and the client side of BioNessieG was implemented through web services. The web services were developed using Apache Axis 2 and Globus Toolkit 4.0.

Globus Toolkit - The Globus Toolkit is an open source software toolkit used for building Grid systems and applications.

It is being developed by the Globus Alliance and many others all over the world. A growing number of projects and companies are using the Globus Toolkit to unlock the potential of Grids for their cause.

The job scheduler was also developed by them which evaluate the suitability and availability of Grid resources before submitting the job to the appropriate computer resources.

BioNessieG end user scenario -

typical scenario for a BioNessieG end user is as follows:

The end user submits a batch of parameter scans (jobs) to the Grid service through the GUI interface.

BioNessieG then divides the parameter scan into several jobs (simulations), each job being represented by a SBML file and saved in a local directory folder BioNessieG.

The web service then checks if the authority of the client who sent the job has sufficient privileges, and if this is the case the jobs are decompressed and distributed to the available resources by the job scheduler.

Each job is then executed and the corresponding results are sent back to the web service in a text file format, which are compressed there and sent back to the BioNessieG client.

BioNessieG decompresses the result file and merges all results files to a single XML scanning result file which can be recognized by BioNessieG. Finally, BioNessieG generates a report either as a plot or a text file.

System Requirements

Contact manufacturer.

Manufacturer

Manufacturer Web Site BioNessie

Price Contact manufacturer.

G6G Abstract Number 20617

G6G Manufacturer Number 104218