CONP
Dissemination of FAIR Datasets and Analysis Pipelines Using the Canadian Open Neuroscience Platform (CONP)
September 1, 2019
Samir Das
Associate Director of Technology
McGill Centre for Integrative Neuroscience
Montreal Neurological Institute
|
|
|
A little preamble...
Dark data
McLeod et al., Lancet, 2014 Ferguson et al., 2014, Nature Neuroscience
$11 million Brain Canada grant
3 year mission starting in 2018
Our goal...
Share data!
|
|
The Plan
CONP technology stack
DATS model
Ethics and data sharing
|
|
|
Data Use Agreements |
Registered Access |
Open Access |
GAP analysis
Datasets are...
- hard to find - and access!
- not generally open.
- not well documented or standardized.
- still difficult for researchers to share.
- are not persistent.
- not reproducible.
Building Principles
Data are distributed.
Governance is distributed.
Portal will provide direct access to metadata.
Metadata exist both at the dataset and file levels.
Data and tools are versioned (with integrity checking).
Privacy regulations are fully respected!
Hybrid model of UI tools and programmatic access (API).
The Data
|
The BigBrain is the 1 TeraByte digitized reconstruction of 7404 hi-res coronal histological sections (20 microns isotropic). |
|
Open iEEG atlas
CONP portal
Computational infrastructure
Boutiques
|
|
CBRAIN
|
|
The Process
Ethics
Curation
Quality Control
|
|
Interoperability
Datalad crawler & extractor
Data publishing
Neurolibre
|
Thank you!Want to contribute data? Do you have any criticisms? Further discussions at Pub Lolec after posters Acknowledgements: Alan Evans, JB Poline, Shawn Brown, Cecile Madjar, Jennifer Mercier-Tremblay, Sanitago Paiva, Dave MacFarlane, Melanie Legault, Xavier Lecours-Boucher, Pierre Rioux, Tristan Glatard, Jason Karamchandani, Derek Lo, Pierre Bellec, Nikola Stikov, Simon Duchense, John Breitner, Judes Poirier, Sylvia Villneuve, Michael Hanke, Yaroslav Halchenko, the LORIS team, the CBRAIN team, PreventAD team, INCF, and many others... |