Mapping between Reactome IDs and Rhea are not that apparent; since we include Rhea in the WP RDF, we could do two things in my opinion:
we could include the Rhea ID (if present) from the Reactome model for metabolic conversions.
update the interaction mapping file to current Rhea data, see if problems still are present, annotate reactome interactions with their ID in GPML2RDF code.
Mapping between Reactome IDs and Rhea are not that apparent; since we include Rhea in the WP RDF, we could do two things in my opinion: