Abstract
Python is emerging as a common scripting language for simulators. This opens up many possibilities for interoperability in the form of analysis, interfaces, and communications between simulators. We report the integration of Python scripting with the Multi-scale Object Oriented Simulation Environment (MOOSE). MOOSE is a general-purpose simulation system for compartmental neuronal models and for models of signaling pathways based on chemical kinetics. We show how the Python-scripting version of MOOSE, PyMOOSE, combines the power of a compiled simulator with the versatility and ease of use of Python. We illustrate this by using Python numerical libraries to analyze MOOSE output online, and by developing a GUI in Python/Qt for a MOOSE simulation. Finally, we build and run a composite neuronal/signaling model that uses both the NEURON and MOOSE numerical engines, and Python as a bridge between the two. Thus PyMOOSE has a high degree of interoperability with analysis routines, with graphical toolkits, and with other simulators.
Highlights
In computational biology there are two approaches to developing a simulation
The goal of this paper is to show how the simulator Multi-scale Object Oriented Simulation Environment (MOOSE; http://moose.ncbs.res.in/, mirrored at http://moose.sourceforge.net/) uses Python to address these issues of interoperability with analysis software, graphical interfaces, and other simulators
We recorded the time-series for the membrane potential during the simulation in a MOOSE table object, which can accumulate a time-series of simulation output (Figure 2A)
Summary
In computational biology there are two approaches to developing a simulation. It is very difficult for others to read such a program and understand how it relates to the targeted biological system. In this context, a model is a well-defined set of equations and parameters that is meant to represent and predict the behavior of a biological system. General simulators lend themselves to declarative, high-level model descriptions that have become important part of scientific interchange in the computational neuroscience and systems biology communities (Beeman and Bower, 2004; Cannon et al, 2007; Goddard et al, 2001; Hucka et al, 2002; http://www.morphml.org/; http://neuroml.org, http://sbml.org).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
