MIASE — Minimum Information About a Simulation Experiment

The Minimum Information About a Simulation Experiment project is a community effort that aims at listing the common set of information any modeller need to provide in order to enable the execution and repeat of a numerical simulation experiment, derived from a given quantitative model. MIASE is designed to help modelers and software tools to echange their simulation settings. Once simulation settings can be exchanged, under format encoding MIASE requirements, together with a given set of models, it will be much easier to reproduce a simulation result using any simulation tool. MIASE is part of the Minimum Information for Biological and Biomedical Investigations (MIBBI) MIASE will take part in the combined CellML SBGN SBO BioPAX MIASE Workshop 2009.

MIASE - Guidelines

The MIASE guidelines describe the Minimum Information needed about a simulation experiment in order to repeat it. They do, for example, contain information about:

Information about the models simulated

MIASE recommends to explicitely define all models used in a simulation by providing a specific name and the source of each model. In order to get a desired simulation result, it is often not sufficient to use models as such. That is why, changes that have to be applied to the model before simulation have to be described. Examples of such changes can be the assignment of a new value (e.g. constant, initial concentration), or the change of a mathematical expression (e.g. using different enzyme kinetics).

simulation settings (type of simulation and the corresponding parameters)

Information about the simulations to run

The simulation tasks to be undertaken in order to complete the simulation experiment need to be specified. Typically, that will involve describing how a simulation procedure has to be applied to a specific model, and in which order. Each simulation can be characterized by certain types of simulation procedures to be run (steady-state, timecourse etc.) and the simulation algorithms used to perform them. The information has to be sufficiently detailed so that no arbitrary choices have to be made when setting up the simulations.

Information about the outputs produced

It is often necessary to define the transformations that have to be applied to the raw output of the simulation tasks, and how to provide the final results. These results can be numerical or graphical. For instance, a model of a periodic process can provide just timecourses showing oscillations; or it can, on the contrary, provide phase diagrams, which are more explicit in describing the relationship between variables. An even more striking example of the necessity for output definitions is the bifurcation diagram.

The Simulation Experiment Description Markup Language (SED-ML)

The SED-ML is a formal representation aimed at encoding requirements defined in the MIASE guidelines. SED-ML is described using UML and will be the basis for data models, for example in XML Schema representation.

The KInetic Simulation Algorithm Ontology (KiSAO)

KiSAO is a classification of simulation algorithms and methods, used for the simulation of quantitative models. It will be used for the description of simulation runs, more precisely for the unambiguous reference to a simulation algorithm used in the simulation settings.

MIASE on sourceforge

You can contribute to MIASE on Sourceforge.