Our primary interest is in understanding transcriptional regulation: how does it work, evolve, and respond to perturbations? To this end, we are leveraging a variety of experimental and computational approaches. In general, we seek to develop principled, biologically informed machine learning approaches to identify causal relationships within transcriptional regulatory networks. As much of our work utilizes novel data, such as global run-on sequencing (GRO-seq), or integrates across diverse large-scale datasets (RNA-seq, ChIP-seq, GRO-seq, ATAC-seq, 3-seq, DNA resequencing, etc), we must frequently develop new computational approaches in order to answer critical biological questions.
Our recent focus has been on enhancers and their transcripts (eRNAs) which we detect and study by leveraging true measures of nascent transcription. We have utilized the GRO-seq protocol to gain insight in the behavior of the tumor suppressor transcription factor p53 and CDK8-Mediator. More recently we have significantly improved both the GRO-seq protocol and developed numerous algorithms for GRO-seq analysis. Our work shows that eRNAs originate from active transcription factor binding sites, allowing us to identify which transcription factors are currently active in any particular cell type. Additionally we have demonstrated the ability to identify the key initial transcriptional responders upon drug treatment using our GRO-seq techniques.
Assays such as chromatin immunoprecipitation (ChIP) coupled with steady state measures of expression (RNA-seq) have dramatically influenced our understanding of transcriptional regulation. Transcriptional regulation is largely governed by the interaction of DNA with DNA-binding proteins, including transcription factors, nucleosomes, histone-modifying enzymes, and basal transcriptional machinery. We have demonstrated that transcription factor binding sites rapidly turn over, even at close evolutionary distances. Therefore, to really understand the mechanisms of regulation requires comparisons at close evolutionary distances, such as between individuals within a species. Likewise, we have shown that the act of transcription itself can have a regulatory effect on nearby sites of transcription.
Aneuploidy, an abnormal chromosome copy number, is observed frequently in nature and contributes significantly to population diversity. It has been associated with antifungal drug resistance in yeast, liver regeneration in humans, and rapid adaptation in experimental evolution studies. Despite its widespread existence, aneuploidy is generally detrimental to eukaryotic cells and strongly co-associates with cell death, developmental defects and cancer. Understanding the molecular mechanisms underlying how aneuploidy impacts a cell will provide critical insights into the costs and benefits of aneuploidy.
When we embark on a scientific collaboration, we focus on projects asking questions that are of particular interest to understanding transcriptional regulation or present particular computational challenges. We work with investigators within the University of Colorado community, at other academic research institutions across the country, and in industry.