Computer-aided peptide sequencing by fast atom bombardment mass spectrometry

This paper proposes an efficient method for computer-aided peptide sequencing based on fast atom bombardment mass spectrometry. The method allows us to analyse peptides much larger than 1000 u within a reasonable cpu time and to have a set of amino acid sequences which are consistent with the observed fast atom bombardment mass spectrum to varying degrees. Another feature is that there is no need to pick sequence ion peaks out of the spectrum; all the ions observed are equally taken into account in the analysis so that one can exhaust the possibilities expected from the whole spectrum. For most samples of penta- to tridecapeptides, parts of an amino acid sequence could be reproduced satisfactorily without prior knowledge of amino acid composition. The degree of success in sequence elucidation depends entirely on the spectrum used in the analysis. When fragment ion peaks are clearly observed, the sequence can be predicted successfully. The result obtained for a given amino acid composition was always satisfactory.