Cleaning up relations

[hgoerzig]

Below is a screenshot taken form the bioportal. I think how dichroism is organised is a bit messy. https://bioportal.bioontology.org/ontologies/PANET/?p=classes&conceptid=http%3A%2F%2Fpurl.org%2Fpan-science%2FPaNET%2FPaNET01137 Just a suggestion how to arrange the subclasses:

• versus polarization
  • versus photon polarization
  • versus circular polarization
    • circular dichroism (which is also a subclass of absorption spectroscopy?)
    • x-ray circular dichroism (which can be a subclass of x-ray probe)
    • magnetic circular dichroism
      • x-ray magnetic circular dichroism (which can be a subclass of x-ray probe)
  • versus linear polarization
    • linear dichroism (which is also a subclass of absorption spectroscopy?)

[spc93]

We can look into this. One issue is that the (very convenient) tree representation doesn't really work with multiple inheritance. Then the object of interest is a graph, which is not so easy to visualize.

A more scientific aspect is that circular dichroism is sometimes used in a more general context, e.g. circular dichroism in diffraction.

[spc93]

I have looked at this and I agree the tree structures are not very clear. However, I'm not sure they ever can be. The really important structure is the underlying graph. Using the visualization tool in bioportal is very helpful here.

Look for the visualization of x-ray magnetic circular dichroism: https://bioportal.bioontology.org/ontologies/PANET/?p=classes&conceptid=http%3A%2F%2Fpurl.org%2Fpan-science%2FPaNET%2FPaNET01137&jump_to_nav=true#visualization

To me this looks fine, but we need to consider the superclasses of each term and see if there any mistakes.

There is an error that needs correcting (issue #53) which affects spectroscopy terms (e.g. dichroism spectroscopy).