About Us
Headed up by Dr. Gregory Barnes, our center is the research arm of the UofL Department of Pediatrics work with autism.
Leading the way in finding new ways to address autism and epilepsy in children through basic science research.
Our basic science research program investigates the genomic determinants of behavior and cognition in human and animal models of autism and epilepsy. We conduct studies in models of autism and epilepsy to aid in the identification of therapeutic drug targets to ameliorate the burden of neurologic disease in these children. Special emphasis is placed on the understanding of the contribution of GABAergic signaling, epileptiform discharges and sleep to epileptogenesis and behavior in the broad Autism Spectrum Disorder populations.
Our Research
We've studied whether CBD can help reduce core autism‑related behaviors. We tested CBD in a well‑established mouse model of autism. After two weeks of treatment, we found that a higher dose of CBD reduced repetitive behaviors and hyperactivity, while a moderate dose improved social interaction. These results suggest that different autism‑related behaviors may respond to different CBD doses. Overall, the findings provide early evidence that CBD could become a promising therapeutic option for addressing social difficulties, repetitive behaviors, and hyperactivity.
How does autism affect the brain’s protective barrier, known as the blood–brain barrier (BBB)? That's what we set out to determine. Autism is often linked with repetitive behaviors, social challenges, and biological changes such as oxidative stress and neuroinflammation. Previous studies of human brain tissue have shown that the BBB may be weakened in autism, but scientists still don’t fully understand why. We are working to uncover what causes this disruption and how it may contribute to autism symptoms.
Using a well‑established mouse model of autism, the team examined key components of the BBB—structures that normally keep harmful substances out of the brain. We found that mice with autism‑like traits had lower levels of important tight‑junction proteins, which are essential for keeping the BBB sealed. At the same time, these mice showed higher levels of inflammation, increased activity of immune‑related brain cells, more oxidative stress, and leakage of albumin, a protein that should not cross a healthy BBB. Together, these findings indicate that the BBB in autism may be more permeable, allowing unwanted substances to enter the brain and potentially contribute to neurological changes.
We are advancing autism research by using artificial intelligence to better understand how differences in the brain relate to specific autism behaviors. In this study, we analyzed MRI brain scans from a large public dataset and paired them with behavioral ratings from the Social Responsiveness Scale, a tool commonly used in clinical evaluations.
By breaking autism down into its core behavioral components—such as social communication, repetitive behaviors, and emotional responsiveness—the team trained multiple AI models to identify which brain regions are most strongly linked to each behavioral domain. The AI system was able to distinguish between typical development and mild, moderate, or severe autism with an impressive 96% accuracy. It also pinpointed specific brain areas, such as regions in the frontal cortex, that differ in function across behavioral groups.
This work demonstrates how AI can help uncover the neurological patterns behind autism and potentially speed up and improve the accuracy of diagnosis.