NLM IRP Seminar Schedule

Abstract:

Evolutionary turnover in the cis-regulatory elements (CREs) of the human genome accounts for more than 90% of the phenotypic and disease associated traits. Some CREs undergo higher rates of substitution and if mutated, may be more likely to result in phenotypic changes. Genomic substrates of novel enhancer activity can be repurposed CREs, transposable elements, or neutral sequences leading to de-novo emergence. We used a deep learning model that’s capable of correlating nucleotide changes to differential enhancer activity and found that a large majority of CREs between humans and our next closest relatives, chimpanzees, have evolved by repurposing regulatory activity from other cell types. Our results highlight a set of predisposed elements that are more suited to regulatory innovation due to their sequence composition of transcription factor binding sites (TFBSs). TFBS enrichment analysis suggests that the repurposed elements do not conform to specific transcription programs. I will discuss results of our analysis that leads us to hypothesize that the repurposed CREs may act as redundant enhancers, are inefficiently integrated into the transcriptional circuitry, and buffer the impact of unfavorable mutations to confer regulatory robustness.