The GI numbers are unique record identifiers that NCBI traditionally paired with accession numbers to designate records in the Nucleotide and Protein databases. Since the beginning of 2016, NCBI has been gradually transitioning to singular use of accession.version identifiers.
If you have a list of GIs that you want to update into accession.version identifiers, you can choose your method depending on the size of your list:
1. For a small number (<1000) of GIs:
2. For up to a few thousand GIs:
3. For large (>10,000) conversion tasks: