Abstract: Despite fragmentation of biology into a myriad separate domains of inquiry from molecular biology to ecology, genetics to plant physiology and bioinformatics to biochemistry, there is simultaneously an increasing drive to integrate these rapidly maturing fields. Comparative genomics takes some steps towards integration by placing protein systems in their biological context and enabling a wider range of insight into a protein, its role in the context of the cell, and the cells' role in the context of the environment. Herein I describe the theory of comparative genomics in Chapter 1, proceeding to apply these concepts to bacterial and archaeal datasets. In Chapter 2 I describe analysis of overall topology and bridges between functional modules of abstract protein interaction networks constructed using comparative genomics techniques. I then apply comparative genomics techniques to phenotypic properties and environmental context. In Chapter 3 I describe the use of phylogenetic profiles to identify a protein exclusive to hyperthermophiles and likely involvement in a novel mechanism of thermal stability, that of intracellular structural disulfide bonds. In Chapter 4 I describe further analysis of phenotype, with emphasis on hyperthermophily, by comparing logical combinations of pairs of protein phylogenetic profiles against phenotype profiles. Further analysis of hyperthermophiles is covered in Chapter 5 with description of identification of a family of hyperthermophilic proteins potentially involved in protein shell formation, informed by studies of salient genomic features of carboxysome operon structure. Finally I return to a more global overview of comparative genomics in Chapter 6 by describing the detection of correlated imitation patterns in large multiple sequence alignments of proteins across many genomes as a tool towards fold prediction. As supplements, I include Appendix A, a reprint of a published article on work I did on the crystal structure of carboxysome shell proteins and Appendix B on work I contributed to on analysis of protein interaction networks of M. tuberculosis. My conclusions rest on the utility of comparative genomics for contextualizing proteins in their organismal and environmental context but note that manual inspection is still necessary to make wide-ranging biological conclusions.