A major challenge of post-genomic biology is understanding the complex networks of interacting genes, proteins and small molecules that give rise to biological form and function. Advances in whole-genome approaches are now enabling us to characterize these networks systematically, using procedures such as the two-hybrid assay and protein co-immunoprecipitation to screen for protein-protein interactions (PPI). Large protein networks are now available for many species like the baker's yeast, worm, fruit fly and the malaria parasite P. falciparum. These data also introduce a number of technical challenges: how to separate true protein-protein interactions from false positives; how to annotate interactions with functional roles; and, ultimately, how to organize large-scale interaction data into models of cellular signaling and machinery. Further, as protein interactions form the backbone of cellular function, they can potentially be used in conjunction with other large-scale data types to get more insights into the functioning of the cell. In this dissertation, I try to address some the above questions that arise during the analysis of protein networks. First, I describe a new method to assign confidence scores to protein interactions derived from large-scale studies. Subsequently, I perform a benchmarking analysis to compare its performance with other existing methods. Next, I extend the network comparison algorithm, NetworkBLAST, to compare protein networks across multiple species. In particular, to elucidate cellular machinery on a global scale, I performed a multiple comparison of the protein-protein interaction networks of m>C. elegans, D. melanogaster and S. cerevisiae. This comparison integrated protein interaction and sequence information to reveal 71 network regions that were conserved across all three species and many exclusive to the metazoans. I then applied this technique to the analysis of the protein network of the malaria pathogen Plasmodium falciparum and showed that its patterns of interaction, like its genome sequence, set it apart from other species. Finally, I integrated the PPI network data with expression Quantitative Loci (eQTL) data in yeast to efficiently interpret them. I present an efficient method, called 'eQTL Electrical Diagrams' (eQED), that integrates eQTLs with protein interaction networks by modeling the two data sets as a wiring diagram of current sources and resistors. eQED achieved a 79% accuracy in recovering a reference set of regulator-target pairs in yeast, which is significantly higher performance than three competing methods. eQED also annotates 368 protein- protein interactions with their directionality of information flow with an accuracy of approximately 75%
