An analytical framework for studying transcriptional regulation

Abstract: The state and behavior of any living cell is controlled by a complex interplay of different regulatory processes, with the regulation of transcription playing a major role. When a cell adapts to a new environment it often does that by modulating gene transcript levels, mainly through changes in transcription factor binding events. Therefore, understanding the transcriptional regulation is vital for many biological research fields ranging from understanding cancer metabolism to metabolic engineering. In this thesis, I present and apply an analytical framework for studying transcriptional regulation in a well-characterized eukaryotic model organism, the yeast S. cerevisiae. The framework is a combination of advanced sequencing methods like Chromatin Immunoprecipitation followed by DNA sequencing (ChIP-seq / ChIP-exo) and Cap Analysis of Gene Expression (CAGE) with bioinformatic approaches. The relative binding location of transcription factors in relation to the transcription start site is important for interpretation, therefore the transcription start sites of all genes active in multiple controlled growth environments were determined using CAGE. To use and analyze the gathered data in a reliable and efficient way a high-quality bioinformatics pipeline was established. After establishing the required analytical framework, I employed it in various projects, all aimed to gain a better understanding of yeast transcriptional regulation. In a detailed study of a single transcription factor, I investigated Leu3, the main regulator of leucine biosynthesis. Here, I was able to show that its binding behavior is affected by the availability of leucine in the media, an adaptive behavior that has not been reported before. Metabolic engineering will be increasingly important to support the needs of our society and in order to help with this, I developed a tool for fine tuning conditional gene expression levels using hybrid promoters. This tool is based on a machine learning approach and can be used to improve productivity in large scale fermentations. In conclusion, this thesis lays the foundation for future large-scale studies of transcriptional regulation in S. cerevisiae and can also serve as a blueprint on how to study it in different organisms.