Repository for development of the neuroscience methods ontology.
Source files that are inputs for generating the ttl files.
The code that generates the ttl.
Identifiers contained herein are currently TEMPORARY PLEASE do not let them into the wild!
Ontology outputs to be loaded into scigraph. nsupper.ttl holds an absurdly lightweight upper ontology nsmethods.ttl will hold the methods concepts nsbeing.ttl will hold entities that are subjects/specimines or tools/reagents nsdatatypes.ttl will hold datatypes that are the outputs of methods
Scigraph configuration yaml files. To build scigraph and load the correct ontologies do the following.
Be sure to replace git/
with the path where you have cloned the repos.
cd git/SciGraph/SciGraph-core
mvn exec:java -Dexec.mainClass="edu.sdsc.scigraph.owlapi.loader.BatchOwlLoader" -Dexec.args="-c git/methodsOntology/scigraph/methodsGraph.yaml"
cd git/SciGraph/SciGraph-services
mvn exec:java -Dexec.mainClass="edu.sdsc.scigraph.services.MainApplication" -Dexec.args="server git/methodsOntology/scigraph/methodsServices.yaml"
The objective of this ontology is to provide a lightweight framework for annotating scientific protocols. To that end our initial approach is to separate the entities and the data that are the inputs and outputs of protocols, from the higher level methods concepts that describe those protocols.
The basic structure for each obo file will be a single parrent is_a hierarchy. Other limited modeling may be added later as needed in order to serve other systems that will directly represent protocols themselves.
To make it easy to distinguish between curated and uncurated terms in raw ttl we will by convention (as in, do not rely on this for code because it will not be enforced systematically) use the following:
UNCURATED:1234567
NOTINANONTOLOGY:1234567
CURATED:fragprefix_1234567
Curie mappings: UH OH
Within the annotation ontology synonyms will simply be text strings that are pulled from another more robust ontology (or other source) that is intended to provide full support for the large diversity of types of synonyms as needed by various users. Synonyms can also be entered directly as strings and we will maintain a list of those synonyms that are missing definitions or require more extensive curation and scholarship.
This will enable us to create a sane workflow that can quickly address practical issues with production performance for scigraph annotations while also supporting the needs of scholarship and textmining.
Code is licensed under the MIT license and ontologies are licensed under CC3.0. Source materials are licensed under CC3.0 or their existing license.
If you would like to contribute changes please fork the repository and submit pull requests. Thanks!