In the biodefense and medical diagnostic fields, MALDI mass spectrometry-based systems are used for rapid characterization of microorganisms generally by detecting and discriminating the highly abundant protein mass-to-charge peaks. It is important that these peaks eventually are identified, but few bacteria have publicly available, annotated genome or proteome from which this identification can be made. This dissertation proposes a method of top-down proteomics using a high-resolution, high mass accuracy analyzer coupled with bioinformatics tools to identify proteins from bacteria with unavailable genome sequences by comparison to protein sequences from closely-related microorganisms. Once these proteins are identified and a link between the unknown target bacteria and the annotated related bacteria is established, phylogenetic trees can be constructed to characterize where the target bacteria relates to other members of the same phylogenetic family.

First, the top-down proteomic approach using an Orbitrap mass analyzer is tested using a well known, well studied single protein.  After this is demonstrated to be successful, the approach is demonstrated on a bacterium without a sequenced genome, only matching proteins from other organisms which are thought to have 100% homology with the proteins studied by the top-down approach.  Finally, the proposed method is changed slightly to be more inclusive and the proteins from two other bacteria without publicly available genomes or proteomes are matched to known proteins that differ in mass and may not be 100% homologous to the proteins of the studied bacteria.  This more inclusive method is shown to also be successful in phylogenetically characterizing the bacteria lacking sequence information.  Furthermore, some of the mass differences are localized to a small window of amino acids and proposed changes are made that increase confidence in identification while lowering the mass difference between the studied protein and the matched, homologous, known protein.