Simple modification of a protein database for mass spectral identification of N‐linked glycopeptides

We describe an algorithm which modifies a protein database such that during a database search deamidation is limited to asparagines strictly contained within the N-glycosylation consensus sequence. The modified database was evaluated using a dataset created from the shotgun proteomic analysis of N-linked glycopeptides from human blood serum. We demonstrate that the application of the modified database eliminates incorrect glycopeptide assignments, reduces the peptide false-discovery rate, and eliminates the need for manual validation of glycopeptide identifications.