UC San Diego

In this dissertation, I provide a generalized framework for the in depth molecular characterization of microorganisms using multi-omic data integration. Next- generation sequencing is rapidly becoming a staple in biological research. However, as data generation becomes more routine, greater attention must be placed on the analytics needed to extract useful information from these datasets. Recent work using multi-omic characterization approaches have revealed that microbial genomes and their organizational features are far more complex than previously thought. Here, I present a multi-omic data integration strategy that updates and expands upon previously implemented workflows that follow four principles to study microbial transcription, translation, and regulation : (1) data generation, (2) data processing, (3) data integration, (4) data analysis. At the core of this integrative workflow is the a complete and accurate reference genome assembly. Chapter 2 discusses updated sequences and gene annotation for Thermotoga maritima and Escherichia coli O157:H7 EDL933 using next-generation sequencing. With the genome sequence revealed, expanded annotation of cellular features is achieved quantitatively using a blend of genome-scale experimental methods and complimentary bioinformatics approaches where the genome serves as a normalizing factor. In Chapter 3 this multifaceted approach is used to elucidate the genome organization of T. maritima and revealed novel insights into its hyperthermophilic lifestyle. The detailed characterization of the T. maritima genome organization was applied in Chapter 4 where the genotype-to-phenotype relationship associated with laboratory evolved cultures were revealed. A genomic region excluded from the original genome sequence was found to be modulated in response to the applied selective pressure, underscoring the importance of the accuracy of the genome sequence. Furthermore, recently developed next-generation approaches were implemented in the multi-omic workflow to provided in vivo, genome-scale data on transcriptional regulation (ChIP-exo, Chapter 5) and protein translation (ribosome profiling, Chapter 6). These assays provide stark improvements compared to previously implemented methodologies with respect to their resolution, signal-to- noise, and cost. They also reveal detailed molecular interactions that previously could not be discerned at genome-scale. Collectively, the workflow utilized here will enable researchers to rapidly and cost-effectively characterize microbial systems beyond the one-dimensional genome annotation and towards complete elucidation of the multidimensional genome organization