| For many years, molecular biologists have studied the genetics of a few model organisms. They have been solving the puzzle of life one gene at a time and one phenotype at a time. However, in recent years, a more complicated picture has emerged. The central dogma of molecular biology has been transformed from a two-step process to an entangled web of regulatory interactions and processes. Cellular mechanisms, such as transcription regulation, post-transcriptional modification and post-translational modifications, significantly complicate the nature of biological problems. Moreover, an explosion in the number of sequenced genomes has rendered single gene studies obsolete as it cannot keep pace with the rapid growth of the sequence repertoire. However, the availability of the same genomic sequences has enabled us to develop systems-level analytical tools. The emergence of microarrays revolutionized the field as it made whole-genome parallel measurements feasible. Microarrays are most commonly used for RNA abundance measurement or genotyping; nevertheless, countless experimental platforms can be developed based on this technology and other whole-genome approaches. Here, I introduce a set of powerful computational/experimental frameworks aimed at revealing the genetic basis of complex phenotypes. First, I describe an information-theoretic approach, called iPAGE, which performs a robust module-level analysis of whole-genome studies. Due to its statistical robustness and unparalleled sensitivity, iPAGE is a recurring theme in all the subsequent approaches discussed in this thesis. We used iPAGE in parallel with FIRE, a platform for discovering cis-regulatory elements, to reveal the regulatory perturbations that underlie the emergence and maintenance of the tumor state. Second, I describe an experimental platform based on fitness profiling and module-level analysis to genetically dissect complex phenotypes in bacteria. Ethanol tolerance, our phenotype of choice in Escherichia coli, has significant implications for commercialization of bioethanol as an environmentally sustainable source of energy. Finally, I introduce a powerful approach called ADAM for global mapping of adaptive mutations in bacteria. We used ADAM to discover adaptive mutations in an ethanol tolerant mutant as well as an evolved strain capable of rapid growth on asparagine as the sole carbon source. The platforms described here are fast, efficient and powerful approaches that significantly accelerate the global genetic dissection of complex phenotypes. |
