The invention relates to a method of peptide sequencing from peptide fragment mass data, wherein a plurality of candidate peptide sequences are determined comprises the steps of: calculating peptide fragment masses, searching a plurality of peak data for masses matching said calculated peptide fragment masses, annotating all permutations of said peak data with amino acid sequences that correspond to the calculated peptide fragment masses, extending said potential sequences to resulting masses with additional matching masses, extending stepwise additions until the resulting masses correspond to parental peptide masses or said parental peptide masses minus the mass of water, and identifying at least one peptide sequence by deleting sequences that can not be extended to endpoints of said parental peptide masses, and deleting identical sequences generated.