In the 2002 publication from the Sanger Institute, the most common BRAF mutation found in melanoma was denoted as V599E. However, majority of publlications now refer to this as V600E. Why the difference? Was there a correction published after 2002?
The difference comes from different reference sequences.
For older, long known proteins, a difference of 1 aa often comes from the fact that a lot of proteins posttranslationally lose the initiator methionine and in the old days, people used to count only the aa of the mature protein (often because the aa sequence was derived by protein sequencing). beta-Globin is a good example where the HbS-Mutation is classically called E6V while according to HGVS nomenclature it actually is p.E7V. You will also find discrepancies in reports of proteins that have a signal peptide. For instance, the classical LDL-receptor literature counts aa without the signal peptide, while in more recent papers p.1 is the initiator methionine.
But here, the reference sequence has changed, just compare NP_004324.1 and NP_004324.2.
The Nature paper is a good example of how mutations should not be reported. The authors did not mention the reference sequence anywhere.
The correct HGVS nomenclature of the mutation using the recent ref seq (NM_004333.4 and NP_004324.2) is p.V600E.
In fact, in the originally populated protein sequence had a (arginine - proline) at position 31, which has been corrected -with an additional amino acid- to be (alanine - glycine - alanine) . Therefore, the position of the mutated amino acid in BRAFV600E shifted from 599 to 600.