An efficient algorithm for sequencing peptides using fast atom bombardment mass spectral data.

An efficient algorithm is described for sequencing peptides from sequence ions appearing in fast atom bombardment (FAB) and FAB tandem mass spectra. The following features are incorporated in the algorithm. The members of the set of sequence ions are represented by all possible combinations of N- and C-terminal fragment ions. From the known N- and C-terminating groups and molecular weight (MW) of the peptide, the sequence ions are mathematically re-expressed as N-terminal residue ions and arranged in ascending order. The peptide sequence is computed, in a stepwise iterative procedure, from the mass differences between the mathematically re-expressed N-terminal residue ions and the predicted peptide subsequences for the neighboring ions of lower mass. These mass differences correspond to combinations of known amino acid residues which have previously been computed and tabulated, based upon the FAB fragmentation rules for peptides. The algorithm was successfully applied to sequence the following peptides from their respective FAB or FAB tandem mass spectrum: decapeptyl (MW 1310), angiotensin II (MW 1045), and two 'unknown' peptides (MW 1227 and 1485, respectively). Two criteria used to predict the correct peptide sequence from among many possibilities are the minimum number of amino acid residues and the maximum fragmentation probability per amino acid residue.