| Description |
COMMITTEE:
Dr. James Galagan, Department of Biomedical Engineering (Major Advisor)
Dr. James Collins, Department of Biomedical Engineering (Chair)
Dr. Daniel Segre, Bioinformatics Program
Dr. Tarjei Mikkelsen, Broad Institute
ABSTRACT:
In order to respond to environmental stimuli, cells utilize an interconnected network of molecules to regulate biological processes. This work describes the application of next-generation sequencing data to unravel these networks, with a focus on ChIP-seq for the identification of transcriptional regulatory networks. A ChIP-seq pipeline designed to analyze high coverage sequence data is described. This pipeline utilizes a simple model of background coverage, along with filtering steps and blind deconvolution to identify sites accurately and at high resolution. Comparisons to other peak calling algorithms on ChIP-seq experiments for previously characterized regulons show that this pipeline is the only method that identifies all previously identified targets for two previously characterized transcription factors in Mycobacterium tuberculosis, DosR and KstR.
This pipeline has been applied to ChIP-seq data in order to identify a genome scale regulatory network of the human pathogen Mycobacterium tuberculosis This network was identified using an inducible promoter system to induce the expression of over half of the transcription factors of Mycobacterium tuberculosis. This network is highly interconnected, containing more binding sites than initially expected for each transcription factor assayed. A subnetwork involving hypoxia and lipid metabolism genes is described using molecular profiling data to support these findings. The pipeline was also applied for the identification of a regulatory subnetwork of clock-regulated light-induced genes in Neurospora crassa. This network is also highly interconnected, and shows complex regulatory feedback onto the core clock genes.
Predicted targets of these sRNAs are consistent with the known function of many of the regulators of the sRNA. Together, these studies have produced resources that can be applied to better understand not only the biology of these organisms, but also the general nature of regulatory networks.
|
