June 28th, 2023 – 2 pm
A data-driven approach to model the mouse brain
Dimitri Rodarie (CREF)
Abstract
The brain is one of the most complex biological structures. Its complexity comes from billions of years of evolution and has been extensively studied over the past century. An approach to unravel the brain is to reconstruct it in-sillico, creating models of small pieces of rodent brain tissue, capable of reproducing the structure and activity of the brain. This can be achieved following a data-driven approach which consists in integrating a formidable amount of data into a coherent spatial framework, an atlas.
Atlases are models that allow the spatial registration of various literature data on the brain and the prediction of structural properties based on this data. Yet, there is great variability in literature results due to inter-subject variability (such as age or sex) but mainly differences in acquisition techniques. We need therefore a strategy to integrate disparate literature data into coherent atlas models of the brain. In this presentation, I will describe a data-driven pipeline to reconstruct atlas models of the mouse brain. These atlases should provide a scaffold of the whole mouse brain in which more detailed reconstructions of the brain can be integrated and have meaningful interactions with the other brain regions. The final outcomes of this pipeline are models capable of replicating the activity of the brain which will be later used to study the effect of neuro-degenerative diseases such as Alzheimer or autism.