1. Cis-regulatory molecular biology. A prerequisite for understanding anything about gene regulatory networks and regulatory genomes is to know their functional elements such as genes and cis-regulatory modules (CRMs). While gene identification has become a routine procedure within a genome project, CRM identification remains formidable challenge due to painstakingly slow experimental methodology and our limited computational ability to predict CRMs. In our lab, we develop ultra-high throughput reporter assay tools for functional annotation and characterization of CRMs by expanding the principles of ‘barcoded’ reporter assays.

Figure 1. Architecture of DNA-tag reporter vectors. Each DNA-tag vector is composed of a basal promoter (BP), a GFP open reading frame, a unique DNA-tag sequence, and a core polyadenylation signal. CRM::tag constructs are built, pooled, and delivered into cells.

2. CRMs in gene expression control system. Multiple CRMs control expression of a gene, yet underlying mechanism of CRM actions remains elusive. While classical textbook examples show that each CRM controls specific aspect of gene expression and that little evidences for functional interaction between CRMs, an increasing number of evidences suggest that some CRMs function in a combinatorial manner. If the combinatorial CRM function were prevalent, it would have a wide impact on the studies of gene regulatory networks, regulatory genomes, and the evolution of gene expression control systems. We measure the prevalence of combinatorial CRM function using experimental and statistical approaches. We also exploit the functional and evolutionary implication of combinatorial CRM function.

Figure 2. Models of CRM function for controlling gene expression.

3. Cis-regulatory function and evolution. The relationship between function and evolution is a fundamental question in molecular evolution. Although earlier studies showed a weak negative correlation between functional importance of protein coding genes and evolutionary rate, the best predictor of protein evolutionary rate is gene expression level complicating delineation of the relationship. In contrast to protein coding genes and functional RNA genes, cis-regulatory elements are not expressed, which makes them ideal for studying the relationship between function and evolution. Cis-regulatory function will be represented as Boolean logic and our high-throughput experimental tools will be used for interspecies comparison of cis-regulatory activities.