Use of mass spectrometry fragmentation patterns of peptides to identify amino acid sequences in databases

Abstract

A method for correlating a peptide fragment mass spectrum with amino acid sequences derived from a database is provided. A peptide is analyzed by a tandem mass spectrometer to yield a peptide fragment mass spectrum. A protein sequence database or a nucleotide sequence database is used to predict one or more fragment spectra for comparison with the experimentally-derived fragment spectrum. In one embodiment, sub-sequences of the sequences found on the database which define a peptide having a mass substantially equal to the mass of the peptide analyzed by the tandem mass spectrometer are identified as candidate sequences. For each candidate sequence, a plurality of fragments of the sequence are identified and the masses and m/z ratios of the fragments are predicted and used to form a predicted mass spectrum. The various predicted mass spectra are compared to the experimentally derived fragment spectrum using a closeness-of-fit measure, preferably calculated with a two-step process, including a calculation of a preliminary score and, for the highest-scoring predicted spectra, calculation of a correlation function.

Dependencies are inferred from the claim text.