We are missing some diseases. In this case, for example, Cardiomyopathy, dilated appears among the evidences, but not among the descriptions. However, it does appear in UniMed's file but it is discarded because of its score.
The score there is the one calculated to accept/discard automatic mappings from UniProt's disease lines to MeSH terms, so we are using it wrongly and discarding manually annotated OMIM cross-references, perfectly valid.
Moreover, every score is already above a threshold (-2.5) set as enough to create these automatic mappings, so we should not exclude any MeSH mapping in the file.
We are missing some diseases. In this case, for example, Cardiomyopathy, dilated appears among the evidences, but not among the descriptions. However, it does appear in UniMed's file but it is discarded because of its score.
The score there is the one calculated to accept/discard automatic mappings from UniProt's disease lines to MeSH terms, so we are using it wrongly and discarding manually annotated OMIM cross-references, perfectly valid.
Moreover, every score is already above a threshold (-2.5) set as enough to create these automatic mappings, so we should not exclude any MeSH mapping in the file.