One of the most profound insights from the genomics era is that most variation among individuals is due not to mutations affecting coding genes but rather mutations affecting how, when, and where genes are expressed. These non-coding mutations are also disproportionately represented among mutations linked to human disease and play an important role in the evolution of new species and adaptation to changing environmental circumstances.
Unfortunately, these mutations are also very difficulty to identify. Each individual carries thousands of non-coding mutations, but only a small fraction of these mutations are functional. Unlike for coding DNA, the relationship between regulatory DNA sequence and regulatory function is quite flexible, making it exceedingly difficult to distinguish functional from non-functional mutations in regulatory DNA.
The fundamental goal of this projects is this: By using recent advances in the ability to profile DNA-regulatory sequence in individual cells, can we efficiently train computational models that distinguish functional from non-functional mutations in regulatory DNA?
Using the sea urchin as our primary model system, the research proposal consists of the following parts:
1) The development of a set of biochemical methods for identifying both regulatory DNA elements and nascent RNA transcription in individual cells (methods that will allow us to develop a training set)
2) Can we use these tools to understand first how cells change their regulatory element usage during changes in cell fate and second can we predict which mutations among individuals and between closely related species are most likely to impact this process?
3) Can we use the computational methods developed in part (2) for predicting the impact of individual, relatively recent mutations to explore very deep evolutionary relationships? Specifically, can we identify similar functioning regulatory elements by computational modeling that might miss similarities between species based on sequence conservation alone?
