Extend model specification to include references to external neuroscience ontology

[salvadord]

Common concepts part of the specification such as “neuron”, “network”, “ion-gate” etc. should make external references to standardized controlled vocabulary concepts. This would allow for unambiguous model interpretation and reuse in accordance with advanced FAIR principles. This is inline with the NetPyNE project roadmap. We aim to facilitate step of searching and converting experimental data to model parameters through the use of knowledge base tools, such as the NIF Discovery portal, the EBRAINS Knowledge Graph (https://kg.ebrains.eu/) or the Blue Brain Nexus (https://bluebrainnexus.io/). Mapping NetPyNE specification components to an ontology will be essential to make use of these knowledge graphs.

Steps to follow:

1. Explore different neuroscience ontologies available and decide more suitable for NetPyNE specification:

• Subcellular Anatomy Ontology https://bioportal.bioontology.org/ontologies/SAO - For example “Neuron” could be defined by the Subcellular Anatomy Ontology concept for Neuron (SAO: 1417703748).

• NIF ontology - https://github.com/SciCrunch/NIF-Ontology

• NeuroLex ontology (used by NeuroML) https://scicrunch.org/scicrunch/interlex/dashboard

• https://neuinfo.org/ is a good reference for neuroscience ontology entities

2. Associate the different components in the NetPyNE specification to the ontology concepts, by formally specifying this in the JSON-based schema (https://github.com/suny-downstate-medical-center/netpyne/blob/development/netpyne/metadata/metadata.py) and the new specification website (https://github.com/suny-downstate-medical-center/netpyne/blob/development/doc/source/modeling-specification-v1.0.rst).