Open ReneRanzinger opened 11 months ago
Raja:
Yes. If possible have a link to FAQ saying what is in GlyGen and why this ID might not be present.
This might be too much work. If there is interest from our users then we should try to use EBI webservices to do this part. I don't think we should do this.
There is one more item which I am not sure should be part of ID mapping. It could be a different tool. It is the software that EBI did for us to map TrEMBL entries to Swiss-Prot when it made sense. Not sure what the blocker is for Jeet. We need to ask him. Many thanks, Raja
Protein
In the ID mapper: If the user provides a valid UniprotKB Accession that is not in GlyGen he currently gets an error message that this is not a valid ID. Now, based on mapping files from UniProt this ID will be mapped to the requested ID space (if the mapping is in the mapping file) and shown as normal result. However, the mapping line will have no “Internal ID” and no link to GlyGen.
Glycan
Similar to above but based on the GlyCosmos mapping for Glycans.
Extension
An extension of (B) that will require us to not only pull the mapping file from UniprotKB but from ANY (XREF) databases and do the mapping based on all of these.