Yang (Eric) Li, PhD, an assistant professor of genetics and neurosurgery at Washington University School of Medicine in St. Louis, has received a five-year Maximizing Investigators’ Research Award (MIRA, R35) from the National Institutes of Health, to develop computational approaches for understanding how gene regulation differs between humans and animal models and how those differences may affect the study of human disease.
Animal models are an essential tool for studying disease and testing potential treatments, but findings in animals do not always translate to humans. Li’s research aims to better understand the genetic and epigenetic differences that may contribute to this gap.
“Many treatments that work well in animal models fail in human clinical trials,” Li said. “We want to understand why.”
Decoding human-specific gene regulation via comparative genomics
Li’s team will integrate single-cell, multimodal genomic data from humans and animal models to match similar cell populations across species and determine where their gene regulatory programs differ. The researchers will then use artificial intelligence and machine learning to decipher the regulatory “code” that controls gene activity.
A major focus will be understanding how regulatory DNA elements, such as enhancers, have evolved between species. These elements help determine when and where genes are turned on or off, and differences in these regulatory programs may help explain why a disease mechanism or treatment response observed in an animal does not always apply to humans.
The work builds on Li’s previous research mapping epigenetic landscapes across more than 3 million individual cell nuclei from human and mouse adult brains. His studies have identified substantial differences in cell populations and gene regulation between species, including differences in regulatory regions associated with human disease risk.
For example, his research found that genetic variants associated with Alzheimer’s disease risk are enriched in human-specific regulatory elements, highlighting the limitations of relying solely on conserved regions between humans and mice.
Using AI/ML to interpret human disease risk
The new R35 will combine comparative genomics with explainable machine learning models to predict functional regulatory elements across species and investigate how regulatory differences relate to disease.
The team also plans to design synthetic regulatory sequences that can be introduced into animal models to help study human genetic risk variants. In some cases, a human risk gene may not be regulated or expressed in the same way in an animal model. Synthetic regulatory elements could potentially help researchers recreate aspects of human gene regulation in those models.
Ultimately, Li aims to create computational tools and knowledge bases that can be applied across different animal models, cell types and diseases.
“We are excited to build a computational framework that can be widely applied to different models and diseases,” Li said.
Flexibility for new discoveries
Unlike a traditional NIH R01 grant, which funds a single defined project, this R35 supports an investigator’s overall research program, giving the lab greater flexibility to pursue new directions as they emerge. For Li, this R35 (R01-equivalent) award represents an important milestone for his research program.
“This is definitely a milestone,” Li said. “It gives my lab the freedom to follow the science and shift directions as new questions arise.”
Li credits his mentors, colleagues and the supportive environment at Washington University School of Medicine for helping him reach this point.
“This year I’m welcoming both this NIH award and a new baby, lots to celebrate!” he said.
Over the next five years, Li looks forward to building his research team and working with young scientists who share his interest in addressing foundational questions in biomedical research.