The simulation base for SANKET uses open standards tools and open interface code to support the entire pipeline of model building, standalone simulations, and driving the models from a database of experiments. The Findsim project specifically interfaces with the database and MOOSE for immediate user access. However, all elements of this pipeline are open and standards based, to facilitate the use of other community tools such as simulators and optimization routines.
MOOSE (The Multiscale Object-Oriented Simulation Environment) is designed to simulate neural systems ranging from subcellular components and biochemical reactions to complex models of single neurons, circuits, and large networks. MOOSE can operate at many levels of detail, from stochastic chemical computations, to multicompartment single-neuron models, to spiking neuron network models. MOOSE uses a Python scripting interface and is therefore easily integrated with Python scientific workflows, such as FindSim.
FindSim is the Framework for Integrating Neuronal Data and Signaling Models. It codifies a standard format in which to define typical signaling and neurophysiology experiments. It provides a core program for running these codified experiments on specified models, to generate a score for how well the model matches the experiment. Using this, FindSim provides additional programs for doing parameter sweeps and optimization.
FindSimWeb is the web interface to FindSim. The project supports a web interface for three key steps: Codifying experimental data into a FindSim- compatible database of experiments; running experiments on models, and (forthcoming) performing optimization on the models so that they better fit the experiments. In addition it provides a management framework for individual and community projects that use FindSim under SANKET.