Working with semantic metadata

Section author: Tommy Yu

PMR2 release 8 and 9 brought in the support for semantic metadata, which allows users to add whatever metadata and annotations they might have stored into the repository into the underlying metadata semantic engine, which then allows them to be retrieved using search queries. In this section, we will go over how to use OpenCOR to annotate a model, and how to add the metadata to the underlying metadata engine then query for the results.

Preparation

In this section, we assume that you have already run through the tutorial on using the repository for the basic operations of the repository.

For the tutorial, we will use a fork of the the Hodgkin, Huxley, 1952 workspace. If you need a quick reminder on how you might do this, please see this section of the tutorial.

Once you forked that workspace, you should now clone that workspace onto your system. If you need help on this, please refer to this help on cloning a workspace.

Using OpenCOR for model annotation

Use OpenCOR to open your local clone of your model file, specifically the hodgkin_huxley_1952.cellml file.

_images/metadata-opencor-start.png

Select the sodium_channel component under the list of components, then click on the helpful link to remove the existing metadata for that node.

In the dropdown menu of Qualifiers, select bio:isVersionOf

In the textbox Term, type in “sodium channel”, as the component is named so. Wait for the possible terms to be retrieved and populated by OpenCOR.

Once that is done, hit the green ‘+’ button for the “sodium channel complex” (GO:0034706) to denote that the component is a version of this term.

_images/metadata-opencor-annotating.png

Now select the potassium_channel component, and repeat the processs to annotate this with the “potassium channel complex” term.

Once you are done, save your changes, commit and push your work back into your private fork. Again, refer to the tutorial linked in the preparation section if you need a primer.

Getting your workspace indexed by the repository

In order to have your annotations pulled into the semantic engine, you need to inform the repository which resources in your workspace should be indexed. Once you have done this, all future versions of the indexed resources will be used to update the semantic engine.

Now that your changes have been pushed back, go back to the page for your fork of the model and select the “RDF Indexing” tab.

_images/metadata-pmr2-rdfindexing.png

Scroll down the left-handed list until you see the hodgkin_huxley_1952.cellml file, select it, then push the button with the right arrow on it to add it onto the list of paths to be tracked, then select the “Apply Changes and Export To RDF Store” button.

Go back to the main page, select the “Ontology based search engine” link at the bottom, then enter the relevant search term. As there are limited reasoning capabilities built into the current iteration of the search engine, you may enter a term one level up above the terms we annotated the model with. For our example, please enter “cation channel complex” into the search box, select the term ending with (GO_0034703). The search indicator will give a green checkmark and now you may select the “Search” button. The search result will now list the workspace and the file that contain this annotation.

_images/metadata-pmr2-search.png