Supporting IMPC needs for mapping MP to GWAS annotations

[dosumis]

EPIC

Background - IMPC would like to be able to mine GWAS data for evidence of phenotypic effects on the genes they target, grouping the results with IMPC phenotype annotations with MP,

e.g.

Strategy: Maximise MP -> OBA and OBA -> EFO measurement mappings for the MP terms used by IMPC

Aims:

• OBA should have equivalent or closely related trait terms for all MP terms used by IMPC to annotate phenotypes (there are about 1000 of these - link to table TBA). The first phase of this mapping should consist of generating OBA terms from MP terms used by IMPC that are patternised or have non-patternised logical defs.
• For the EFO measurement : OBA mappings, we should prioritise cases where the lexical breakdown matches E & Q in MP terms used by IMPC.

Notes

1. IMPC also has many free text measurement terms (parameter name) - upstream of MP assignment. Unless there is a strong use-case for it, we will not attempt mapping of these to OBA, and even if there is, this is a lower priority aim than MP -> OBA -> EFO measurement.
2. IMPC also uses anatomy terms in annotation - MA and EMAP. These can be mapped into Uberon if needed, but this also needs justification by use case.

[matentzn]

Using approach we discussed today: https://github.com/obophenotype/bio-attribute-ontology/issues/176