A "bring to MINDS" approach for large scale neuroscience knowledge graph

Talk Title

Session Overview

One of the key step of the EPFL Blue Brain Project data-driven approach to digitally building and simulating the mouse brain involves acquiring, organising, and integrating heterogenous amounts of data from many sources including neuroscience experiments, published scientific papers, and brain databases. The data describe the structural and functional organisation of the brain at various levels and is used to build brain tissue models which can then be disseminated to the neuroscience community. To address many of the challenges raised alongside this iterative cycle, BBP developed and deployed Blue Brain Nexus (https://bluebrainnexus.io/, https://doi.org/10.3233/SW-222974), a data and knowledge graph management system leveraging W3C open and interoperable standards such as RDF and W3C SHACL. In this presentation, I'll talk about how a large-scale knowledge graph is used at Blue Brain by decentralised domain teams to go beyond data cataloging. High-quality metadata using ontologies and SHACL schemas are used to handle the diversity and evolution of data at the scale of the whole brain, on the one hand, to track the data provenance as a support for quality assessment, reproducibility and accurate attribution on the other. Finally, the ontologies and graph structure can be leveraged to support content based entity recommendation and entity linking.

Target Audience

data scientists, knowledge engineers, technical practitioners, data architects

Desired Feedback

What kind of feedback are you seeking from the audience? Are there specific questions or challenges you want to address?

Speaker Details

Mohameth François Sy, Data and Knowledge Engineering manager at EPFL Blue Brain Project (https://mfsy.github.io, https://www.epfl.ch/research/domains/bluebrain/blue-brain/people/group-leaders/mohameth-francois-sy/)

Recording Consent

YES

zazuko / knowledge-graph-forum