Why does the protein position in the protein_change annotation differ by one from UniProt?
This happens for certain GENCODE v19 transcripts. These transcripts had a missing base and the protein sequence had a single amino acid dropped. Hence, the protein position in the protein change looks to be off by one. If you use the latest uniprot_aa_xform datasource, the UniProt_AAPos annotation will have the corresponding position in UniProt