Searching Sequence Databases via De Novo Peptide Sequencing by Tandem Mass Spectrometry

Abstract

The initial stages of tandem mass spectral data interpretation for a pro-teomics project usually involve looking for exact matches between the query spectrum and theoretical fragmentation data derived from database sequences (1–3). A number of programs are available that use the observed peptide mass as a filter for identifying portions of database sequences with the same calculated mass prior to evaluating the congruency between the database sequence candidates and the tandem mass spectrum (4–7). These programs generally fall short when analyzing peptides whose sequences differ in a manner that alters its mass compared with the database sequence (see Note 1 ). For example, a single conservative substitution of isoleucine for valine is sufficient to thwart algorithms that rely on peptide mass prefilters. We commonly obtain high-quality tandem mass spectral data of peptides for which no exact database match can be made, and the question remains whether these nonmatching spectra are due to novel sequences, or are a result of less interesting possibilities such as interspecies variation, database sequence errors, or unexpected pro-teolytic cleavages.