Steven Reilly, Ph.D.

Research

With my research group, we will design programmable, synthetic cis-regulatory elements—DNA sequences that control gene expression—to facilitate the delivery and activation of therapeutic cargos to specific target tissues. These regulatory elements dictate when and in which cells or tissues human genes are switched on, but while the potential sequence combinations in even a short DNA sequence are vast, evolution has only explored relatively natural elements. Understanding exactly how DNA encodes this regulatory logic and grammar—and how genetic variation in these elements’ operation can lead to disease—is an ongoing challenge. Our team developed a system that combines massively parallel assays of gene regulatory activity across cell types and machine learning methods to predict regulatory function from DNA sequences alone. Combined with generative artificial intelligence, we can design synthetic regulatory elements, never seen before in nature, with programmable functions such as specific levels of gene expression or tissue specificity. In my lab, we will engineer and characterize thousands of synthetic regulatory elements that can drive gene expression in major brain cell types—astrocytes, microglia, and neurons—while remaining inactive in nontarget tissues, such as the liver and spleen. We will assess if these highly specific regulatory elements can improve targeted delivery of genetic therapies for Alzheimer’s disease or Parkinson’s disease in preclinical models. The work could provide an important step toward realizing the promise of gene therapy and personalized medicine.