Neuroinformatics, computational neuroscience and artificial intelligence.
The recent development of experimental technologies raises many challenges to neuroscience research: multiscale data (from gene and cell to individual), with a strong heterogeneity (electrophysiological data, images, verbatim, behavioral data...), voluminous and of complex nature. Combining these data is both a scientific and methodological lock which need to be broken to better understand normal and pathological functioning of the human brain. A classical monodisciplinary approach is no longer sufficient and interdisciplinary works combining experimental approaches, modeling and multiscale computer analyses are better suited to tackle these problems with their whole complexity.
Within this framework new methodological and research fields emerged, as the neuroinformatics, aiming to use neuroscience data both as an object of study and as a source of inspiration to imagine new analysis methods, in particular in the artificial intelligence field (AI).
At CRNL, this axis aims to develop and improve mastering of mathematics and informatics tools helping our research work.
More specifically, it aims to promote: (i) knowledge sharing and training in existing methods, (ii) supporting our innovation in computational neuroscience and AI application to our research questions.
This axis opens two complementary perspectives:
-> broadcasting methods for analysis and synthesis of data
-> conceiving models of studied phenomenon (either descriptives or mechanistics or normatives)
Many actions of this axis favor a regular and continuous exchange between implicated CRNL members.
A few specific main events (international workshop, conference satellites) allow to open more broadly the CRNL to national and international communities working in this domain.
- Methods club (weekly meeting where we answer questions and needs, exchange and get trained together)
- Yearly trainings: coding (Python/Matlab), softwares (MNE/R)
- International or local collaborations, in particular a participation to the IXXI institute (Institute for complex systems in Rhône-Alpes)
- We are involved in teaching to M1 students (optional module in computational neurosciences)
- Hiring of students, engineers, researchers and teachers skilled with theoretical or informatics tools required for computational neurosciences.
- To continue to promote methods from neuroinformatics or computational neuroscience fields in the CRNL, in particular through seminars or workshops aiming to demonstrate their utilities and to explain how to use them.
- To favor the share of tools in the laboratory (code sharing, data sharing)
- To pursue the development of exchanges between members on this topic (methods club, Slack, wiki, trainings)
- To improve initial training offer (from L3 up to M2 with a dedicated cursus)
Moustafa Bensafi (NEUROPOP) : moustafa.bensafi@cnrs.fr
Nicolas Fourcaud-Trocmé (CMO) : nicolas.fourcaud-trocme@cnrs.fr
Jérémie Mattout (COPHY) : jeremie.mattout@inserm.fr