PLATO: Platform for integrating large-scale brain model from sub-system models

Takayuki Kannon (RIKEN Brain science institute), Keiichiro Inagaki (Computational Science Research Program, RIKEN), Nilton Kamiji (RIKEN Brain science institute), Kouji Makimura (RIKEN Brain science institute), Shiro Usui (RIKEN Brain science institute)

Most mathematical models in neuroscience have focused on specific sub-systems such as functional areas in the brain. In order to understand details of brain function, each sub-system model needs to be integrated toward the large-scale brain system model.

In the present work, we propose a data-oriented model integration method. The main idea of our approach is to commonize the input and output of each model, so that any kind of models can be connected to each other. Our aim is to provide an integrated development environment platform for large-scale model assembled from sub-system models by means of a common data format. This platform includes the data-oriented model integration library named PLATO Network Interface Class Library (PLATONIC), a graphical user interface editor, and a database that manages common data used for model integration. This platform enable us to reuse and to integrate newly developed and existing sub-system models by minimum changes in the program codes towards development and simulation of a large-scale brain system model. As an example we here demonstrate a large-scale visual system model which covers eye optics, retina, several function of visual cortex and eye movement control to study visual illusion.


This work is partially funded by "The-Next Generation Integrated Simulation of Living Matter" project, part of the Development and Use of the Next-Generation Supercomputer Project of the Ministry of Education, Culture, Sports, Science and Technology of Japan.

Preferred presentation format: Poster
Topic: Large scale modeling

Document Actions