Data pipelines

Data Pipelines

For our social pair task, we want to build a data pipeline rather than the usual ad-hoc data storage.

Here are some links. The idea is to use either Alyx + DataJoint or Datajoint alone and follow the NWB specs as much as possible:

• Datajoint: https://datajoint.com/docs/core/
• Alyx: https://alyx.readthedocs.io/en/latest/
• Simpler alternative: Expipe - https://github.com/CINPLA/expipe
• DataJoint for MATLAB
• NWB Core Principles Paper

This paper for a recent update on IBL and its solutions (alyx etc.): A modular architecture for organizing, processing and sharing neurophysiology data - Nature Methods 2023

fig

Other links:

• https://neurosciencecommunity.nature.com/posts/data-architecture-for-collaborative-neuroscience
• https://open-neuroscience.com
• https://www.sciencedirect.com/science/article/pii/S2772528622000346