Improving epidemiological surveillance algorithms for Francisella tularensis using mass spectrometry and proteomics

This dissertation investigates the genomic and proteomic characterization of Francisella tularensis isolates using advanced bioinformatics tools and mass spectrometry techniques to enhance pathogen identification, epidemiological surveillance, and public health response. Whole genome sequencing (WGS) was employed alongside tools such as CanSNPer2, CARD, and RAxML to classify F. tularensis canSNPs, detect antimicrobial resistance, and understand the genetic diversity of isolates primarily collected from Oklahoma and North Carolina. The study achieved a 95% classification accuracy in determining canSNPs, with antibiotic resistance genes vanYB and FTU-1 identified across all isolates. These findings underscore the utility of WGS in microbial forensics and outbreak management. Additionally, this research explored the application of Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) for comprehensive proteomic analysis of F. tularensis isolates. LC-MS/MS identified differential protein expression between subspecies, with proteins such as isochorismatase-like domain-containing protein and indolepyruvate decarboxylase showing higher expression in F. tularensis subsp. tularensis. The study also demonstrated that machine learning models, specifically support vector machines (SVM), could classify F. tularensis isolates with 98% accuracy using proteomic data, highlighting the potential of LC-MS/MS in pathogen classification. Furthermore, phenotypic analysis using Biolog Omnilog revealed metabolic diversity across isolates, emphasizing the value of combining genomic and phenotypic data for a comprehensive understanding of bacterial behavior. This research contributes to public health and biodefense strategies by providing rapid identification tools for bacterial biothreat agents, improving pathogen detection and classification in clinical and field settings. The findings have significant implications for future research in bacterial genomics, proteomics, and epidemiology, suggesting that LC-MS/MS is a promising tool for enhancing the accuracy of microbial diagnostics.