Including mutations from conceptually translated expressed sequence tags into orthologous proteins improves the preliminary assignment of peptide mass fingerprints on non-model genomes

Abstract

In order to improve protein assignment from peptide mass fingerprints (PMF) in species with incompletely sequenced genomes, the genus-specific mutations deduced from Expressed sequence tag (EST) sequences were included in the complete reading frames of orthologous proteins, resulting in a new searchable in silico protein database. Using this method in tests on four plant species, the MOWSE score of at least 20% more proteins was improved compared to conventional approaches on crude, total proteins, for middle-sized EST projects. Larger contigs are assembled in more important EST projects and this improves the conventional assignment of the most abundant proteins. However, contigs from minor transcripts remain shorter and the assignment of less abundant proteins, such as those isolated following subcellular fractionment, is improved by searching orthologue-EST conceptual chimeras with the PMF spectra. This strategy may be utilized as a tool to identify potential PMF matches that can be then verified by other experimental approaches (tandem mass spectrometry) to ensure the EST matched chimera identification is accurate.