Open cmungall opened 6 years ago
location is very much tied to geographical addresses.
From memory, I think the motivation was to allow location to be used in two different ways.
Personally I think that we should separate out these two uses as it leads to the confusion that is highlighted in this issue.
We are now proposing to split location in a structure from geographic location. The proposal is to use a new property for the location in a structure.
This no longer relates to BioChemEntity but to SequenceRange
I'm trying to understand the location model in bioschemas
In http://bioschemas.org/types/BioChemEntity/
I see we have a 'location' field, the values of which may be Place OR PostalAddress OR PropertyValue OR Text OR URL
I'm not sure what the use case for allowing postal addresses on genes is?
The notes says:
There is a bit ambiguous - exactly what FALDO class is expected here?
I can only find one example:
https://github.com/BioSchemas/specifications/blob/4e666e07f1c2c6ed0cacac81fb661d67a31f22b0/BioChemEntity/examples/example.json#L42-L46
The gene has position 837 - what does this mean?
I would recommend having this be more formally specified, with a BioChemEntity being equivalent to or a superclass of SO types, and faldo:location being mapped to bs:location, with a faldo:Region being the range.
cc @JervenBolleman