Add all possible annotation and annotation extension relations to Noctua form

[vanaukenk]

Following up from 2022-03-02 manager's call where we discussed adding all possible annotation and annotation extension relations to the Noctua form, i.e. adding all relations currently available to curators in other tools or Noctua workbenches.

@thomaspd below is the list gathered from Protein2GO (note that below is only the subset of relations in Protein2GO that are GO-CAM applicable) and ShEx. Do you have time to review this together? Thx.

BP relations:
part of
has input
has primary input
has output
has primary output
occurs in
has target end location
has target start location
causally upstream of
causally upstream of negative effect
causally upstream of positive effect
regulates
negatively regulates
positively regulates
results in development of
results in division of
results in formation of
results in morphogenesis of
results in acquisition of features of
results in commitment to
results in determination of
results in specification of
results in assembly of
results in disassembly of
results in organization of
results in maturation of
results in growth of
results in movement of
results in remodeling of
transports or maintains localization of

MF relations:
part of
has part
occurs in
has input
has primary input
has output
has primary output
directly provides input for
regulates
negatively regulates
positively regulates
directly regulates
directly negatively regulates
directly positively regulates
causally upstream of or within
causally upstream of or within positive effect
causally upstream of or within negative effect
causally upstream of
causally upstream of negative effect
causally upstream of positive effect
happens during
transports of maintains localization of
has target start location
has target end location

CC relations:
part of
adjacent to
overlaps
existence overlaps
existence starts and ends during

PCC relations:
located in
is active in
has part
part of
existence overlaps
existence starts and ends during

GP relations:
contributes to
located in
part of
colocalizes with

[pgaudet]

I thought we wanted to lint to the relations allowed in ShEx, and remove 'leftover' relations that we haven't been able to fix yet, ie 'directly regulates'. What about the the chained relations?

[vanaukenk]

In GO-CAM, there is an alternative way to model the biology captured by chained relations, so we don't absolutely need them. Conceptually, we won't lose any information, it will just be annotated a bit differently.

For other relations, yes, we have said we would phase some of these out, but this is currently what's in the ShEx and in Protein2GO. We can continue to move forward with our decision to remove some of these (e.g. directly regulates) but I wasn't entirely clear what the relations project status was for dealing with these others. We still need to coordinate the efforts here across tools and other software, so someone needs to be responsible for doing that.

[vanaukenk]

Here's a summary table with AE relations and their current status wrt Noctua and ShEx.

[vanaukenk]

From 2022-12-01 workbenches meeting:

We decided that if a more specific BP term shares a relation with a relation in the root BP shape, we will override displaying the relation and its constraints from the root BP shape and instead display the more relation and constraints from the more specific BP term.

See for example: Metabolic Process