Discovering Regulatory Insights from Gene Expression Dynamics


Cells are complex, autonomous machines that integrate many environmental cues to execute a desired response. Though this property makes cells versatile, it presents significant design challenges when, to treat diseases, we must alter cellular responses. To understand changes to the complex regulatory pathways that cause diseases, studies often investigate the differential gene expression between genetically or chemically differing cell populations. This approach transformed the discovery of genetic drivers of disease and possible therapies. Current high-throughput technologies also provide a wealth of time-series data that captures complex regulatory dynamics, yet many current analyses do not capitalize on this temporal information to provide quantitative predictions of gene expression in untested conditions. A better understanding of gene expression dynamics will lead to more detailed and quantitative models of cellular regulation. This improvement can accelerate our understanding of biological systems, guide future experiments, and enhance our ability to control cellular behavior. In the following work, I present two distinct approaches that utilize gene expression dynamics to elucidate systems that regulate transcription. I developed algorithms to identify genes that regulate each others' expression, create dynamical systems models of expression that accurately predict gene expression in multiple contexts, and gain insight into the regulators of specific transcriptional responses. I validated my algorithms on in silico and in vitro data. I demonstrate how these techniques revealed unique insights into transcriptional regulation by PI3K and Sprouty. My work illustrates how the principled use of temporal information can improve our understanding of biological systems, and I hope it encourages others to collect more time-series data in the future.

Date created
Resource type
Rights statement