A new approach that allows identification of intron-split peptides from mass spectrometric data in genomic databases

We present a new approach that allows the identification of intron-split peptides from mass spectrometric data in genomic databases. Our algorithm uses small regions of peptide sequence information which are automatically deduced from de novo amino acid sequence predictions together with the molecular mass information of the precursor ion. The sequence predictions are based on selected collision-induced mass spectrometric fragmentation spectra. Fragments of the predicted amino acid sequence are aligned with each of the six frames of the translated genome and the precursor mass information is used to assemble the corresponding tryptic peptides using the sequence as a matrix. Hereby, intron-split peptides can be gathered and in turn verified by mass spectrometric data interpretation tools such as Sequest.