Auto PR: assertion_evidence.yaml generated from google spreadsheets using schemasheets

[puja-trivedi]

This PR adds a autogenerated assertion_evidence.yaml file.

[djarecka]

I have several questions:

• @tekrajchhetri - I remember that we discuss that I would like to have connection between Agent and both Assertion and EvidenceLine, but perhaps in that case we should have the same type of connection with Agent and Activity. Right now EvidenceLine has a direct connection to Activity by output_of and Assertion does not (although it could have indirect through Agent)
• can someone explain issue sepio vs dcterms and why we are using both?
• I think I saw significance_level and significance_type on sepio website. do we have to create a new brainkb slots? Also they do not have descriptions, we should add it.
• I see that some slots do not have has prefix. Should we standardize it?
• we should have description where the value sets are taken from (if they are not invented by us)
• I would really want to see descriptions for the values in value sets. For Category is kind of clear, but for others it is not
• I don't think Category is a good name, we also use for something else, perhaps we could add a second part to this name
• I see that ContributorTypes are taken from datacite, but I don't think we should just use it without adjustment. It doesn't fit our case completely
• We should add the slots for Annotation

[tekrajchhetri]

@djarecka please find my answers below.

• @tekrajchhetri - I remember that we discuss that I would like to have connection between Agent and both Assertion and EvidenceLine, but perhaps in that case we should have the same type of connection with Agent and Activity. Right now EvidenceLine has a direct connection to Activity by output_of and Assertion does not (although it could have indirect through Agent)

  --> I am not sure if it's appropriate to have connection between Activity and Assertion with output_of relation at the moment. Because I couldn't think of the activity examples that would relate to assertion. Could you perhaps give an example of the Activity?

• can someone explain issue sepio vs dcterms and why we are using both?

  --> Because there are terms that we are using, e.g., dcterms:description and dcterms:source that are not available in sepio. The description in sepio also uses the dcterms. If you've suggestions on particular classes/properties we could change it accordingly. Also, I do not see any issue using both. Is there any issue with this, could you let us know the issue?

• I think I saw significance_level and significance_type on sepio website. do we have to create a new brainkb slots?
  Also they do not have descriptions, we should add it.

  --> I am not able to find it on sepio ontology as you suggested. Could you share the SEPIO full IRI for these properties?

• I see that some slots do not have has prefix. Should we standardize it?

  --> Yes, we should unless we are reusing from somewhere else.

• we should have description where the value sets are taken from (if they are not invented by us)

  --> we should have description. But if they're taken from somewhere else, it's marked by their prefix.

• I would really want to see descriptions for the values in value sets. For Category is kind of clear, but for others it is not

  --> we have descriptions in the google sheet (classes & definitions tab). @puja-trivedi is it not taking descriptions from classes & definitions tab.

• I don't think Category is a good name, we also use for something else, perhaps we could add a second part to this name

  --> that is the best name I could think of. I am open to any suggestions.

• I see that ContributorTypes are taken from datacite, but I don't think we should just use it without adjustment. It doesn't fit our case completely

  --> Yes, we should.

• We should add the slots for Annotation

  --> Do you have suggestions for slots?

[djarecka]

we're meeting, so will not be answering here for everything. But this is the sepio website I'm referring to:
https://github.com/monarch-initiative/SEPIO-ontology/wiki/SEPIO-Overview#the-sepio-information-model-and-profiles

[satra]

why output_of and not generatedBy from prov? we should use standard prov properties unless there are good reasons to create new properties. prov should still be the base model driving our work till someone shows something better.

and regarding activity isn't any model or human based extraction an activity, or at the very least captures when an assertion was made (start/end are properties of activities)

[djarecka]

worked moved to #130