Create parallel hierarchy in RO for causal statement branch of PSI-MI

[cmungall]

https://www.ebi.ac.uk/ols/ontologies/mi/terms?iri=http%3A%2F%2Fpurl.obolibrary.org%2Fobo%2FMI_2234

Note we cannot import this as a branch in RO as MI is classes and RO represents the relation.

Proposal: shadow the causal statement branch, and formally define relationship between the two (very roughly, for a pair (c,r) :if i is an instance of MI class c, then r(i.interactorA, i.interactorB))

cc @rlovering @vtoure @luanalicata @liviap

[vtoure]

Here is the link to a google doc with an attempt to map the current RO and MI terms: https://goo.gl/S1N9qm. Potential conflicts are highlighted.

[dosumis]

The primary relations in RO are process to process (process is used broadly here - covering GO MF). These are under causal relation between processes

This branch has relations for direct and indirect causal interactions.

IIRC MI terms refer to causal relationships between gene products. As noted in the linked doc, these map to causal relation between material entities. This branch lacks relations for indirect regulation, but these would be trivial to add:

activity regulates activity of: Holds between molecular entities A and B where A regulates a process that B is capable of. For example, A and B may be gene products the activity of A regulates the kinase activity of B. This holds for direct regulation (via physical interaction between A and B) and for indirect (e.g. where A is upstream of B in some chain of activities that regulate each other). property_chain: enables o regulates o enabled_by -> activity regulates activity of

Alignment also requires adding relations for up_regulates_quantity of.

Question: Is it safe to assume that the overall activity is being upregulated by upregulating levels?

If so, I suggest we add a new relation under regulates to cover regulation of a process by regulating levels of its effector (example - regulating overall levels of activity of a gene product in a cell by regulating the quantity of that gene product). We should then add the equivalent under causal relation between material entities.

[dosumis]

for a pair (c,r) :if i is an instance of MI class c, then r(i.interactorA, i.interactorB))

Looks easy enough to add as SWRL rule.

[github-actions[bot]]

This issue has not seen any activity in the past 2 years. It will be closed automatically 60 days from now if no action is taken.