Open matthiaskoenig opened 8 years ago
That makes sense to me.
The source for our CHEBI IDs (http://www.metanetx.org/) does not differentiate primary vs secondary CHEBIs, so we just need to check against CHEBI for each one during loading.
Hi all, any progress on this issue? This is very easy to fix. Just check for every chebi id what is the primary chebi id and reduce the annotations to this subset. Best Matthias
Most of the species contain annotations to the secondary ChEBI metabolites. For instance in model
iAB_RBC_283
.Herin, all the the following chebi annotations are secondary CHEBI:14377, CHEBI:11675, CHEBI:39915, CHEBI:44955, CHEBI:44957, CHEBI:1327, CHEBI:19829 and forward to the primary ChEBI 16356: http://www.ebi.ac.uk/chebi/searchId.do?chebiId=CHEBI:16356
see for instance http://www.ebi.ac.uk/chebi/searchId.do?chebiId=CHEBI:14377
The secondary annotations do not provide any additional information but create problems with many services working with ChEBI identifiers, which most of the time work with the subset of primary annotations. For instance the ontology lookup service OLS cannot retrieve information for secondary ids see http://www.ebi.ac.uk/ols/api/ontologies/CHEBI/terms?obo_id=CHEBI:16356 vs http://www.ebi.ac.uk/ols/api/ontologies/CHEBI/terms?obo_id=CHEBI:14377 which is not resolving.
In my opinion the secondary chebis should be dropped from the model annotations. They
The best Matthias