Brain Simulation across Scales
EBRAINS integrated workflows at multiple scales allow users to perform complex computational experiments, including estimation of model parameters, model validation and large-scale simulations including analysis and visualisation. Workflows addressing the needs of scientists across disciplines and levels of expertise will be accessible through pre-configured web applications or flexible digital notebooks.
Building on the conceptual separation between mathematical models on the one hand and simulation and analysis tools on the other (Neuron 2019, 102:735, doi: 10.1016/j.neuron.2019.03.027), EBRAINS provides sophisticated and mature tools and services to explore, modify, create and share brain models, embedded into integrated modelling and simulation workflows. Accessible through pre-configured web applications, flexible digital notebooks, and command line interfaces, the workflows address the needs of scientists across disciplines and levels of expertise, as well as the classroom. EBRAINS overcomes the complexity barrier in brain modelling by digitisation of the full modelling process and enabling the use of models as building blocks for larger ones. In addition, EBRAINS supports multi-scale modelling and co-simulation where a model at a coarse level of description provides the embedding for a more detailed one.
EBRAINS supports tasks, from describing signalling processes in postsynaptic transmission, to exploring a detailed single cell model to help with the interpretation of in vitro recordings via simulations, to the exploration of mechanisms of behavioural responses and cognition, to investigating simplified full-brain models under in vivo conditions. Validation, analysis, and visualisation tools are fully integrated, such that the complete scientific cycle of iterative comparison to experimental data and model refinement is digitally mapped and reproducible. The workflows are transparently connected to HPC systems creating a world-wide unique scientific instrument. EBRAINS supports users in documenting and sharing their work according to FAIR and RRI principles, thus promoting collaborative research and the open publication of models as building blocks for further studies by others.