Image
DIBS is pleased to announce the 2026 cohort of Germinator Award recipients. The graduate students and postdoctoral fellows listed below were selected based on their innovative proposals for interdisciplinary research in the brain sciences. The projects are funded for 1 year at $25,000.
Community engagement in the awards process was extremely high. We thank everyone who participated. DIBS received over 50 letters of intent from applicants across 26 Departments and 5 Schools at Duke. Half of those trainees were invited to submit full proposals, which were evaluated by 25 expert faculty reviewers. We are proud not only of the 6 awardees but also every student and postdoc who submitted their ideas for consideration.
Julia Dziabis, PhD will investigate how viruses hide in neurons to emerge years later and contribute to brain aging and neurodegeneration
Postdoctoral Fellow, Dept. of Cell Biology, Scott-Hewitt Lab
“Neuroimmune biomolecular condensates as intracellular organizers of latent viruses within neurons”
Summary:
Exposure to viruses is a part of life. Some viruses, such as the varicella-zoster virus that causes chickenpox, can “hide” within nerves and remain dormant and undetectable to the host immune system for decades. Chickenpox, for example, can reactivate later in life as shingles, in response to stress or when the immune system weakens with age.
Recent evidence has linked the presence and reactivation of these viruses to conditions such as Alzheimer’s disease and multiple sclerosis.
While a lot is known about the genetic mechanisms that viruses use to avoid the immune system, little is known about where latent viruses physically hide within the nervous system or how their location influences their reactivation. Our lab has recently discovered a novel type of intracellular compartment found only in neurons. This project will test how these compartments help viruses enter latency and later reactivate, using advanced techniques in microscopy, molecular biology, and immunology. The results will offer new insight into how chronic viral infections may contribute to brain aging and neurodegeneration.
Kyle McPherson will work to discover treatments for Age-Related Macular Degeneration (AMD) that preserve both vision and brain health
Image
PhD Candidate, Dept. of Ophthalmology, Bowes Rickman Lab
“Ocular nuclear and transcriptomic heterogeneity preceding neurodegeneration”
Summary:
Cognition naturally decreases as we age, leading to slower information processing and declining memory and attention. For people with worsening vision or blindness, cognitive decline is worse. Vision loss not only damages areas of our brains that process what we see but also affects parts of the brain that allow us to read, speak and remember. Therefore, we need to find ways to preserve sight for the sake of brain health.
1 in 3 people aged 80 years old will develop a progressive form of blindness called ‘Age-related Macular Degeneration’ (AMD). People with AMD progress from mild symptoms to the complete loss of vision. Before this vision loss, ‘support cells’ in the eye become dysfunctional and change shape. By using knowledge from ophthalmology, immunology, and bioinformatics we will address this knowledge gap to better understand the disease changes and develop better treatments preserve both vision and brain health.
Uros Topalovic, PhD will use data from brain recordings, body sensors, and virtual reality to help people with epilepsy and other brain disorders optimize their learning
Image
Post-doctoral Scholar, Dept. of Neurosurgery, Suthana Lab
“Adaptive human neurotechnology platform for cognitive studies”
Summary:
People with epilepsy and related brain disorders sometimes receive implanted devices that continuously record their brain activity. These recordings, together with patients’ reports, suggest that there are “good moments” for when the brain is ready to learn and perform. We will use virtual reality (VR) to take advantage of those moments to improve performance.
To do this, we will build a specialized platform that connects implanted devices, body sensors, and VR in real time. The software will be integrated into a small body-worn device that records brain activity, eye movements, motion, heart rate, etc., to drive a VR training environment that will select the “good moments” to deliver the appropriate information. This platform will support future interdisciplinary studies to test whether the “good moments” can strengthen memory and performance.
Andrea Jones, PhD will examine how estrogen promotes recovery after stroke to guide the development of new treatments that mimic its benefits
Image
Postdoctoral Associate, Dept. of Biomedical Engineering, Segura Lab
“Mechanisms of non-canonical estrogen receptor activation in modulating neuro-regeneration after stroke”
Summary:
Stroke is a leading cause of disability, and recovery becomes harder with age.
One reason is that older adults, especially women after menopause, lose the hormone estrogen, which normally protects blood vessels and supports healing. Women who have higher circulating estrogen levels tend to experience less severe strokes and recover better, but scientists don’t yet understand why.
This project investigates how estrogen helps the brain heal after stroke and what happens when it’s missing. Using a mouse model that mimics menopause, I will study how the absence of estrogen changes brain inflammation and repair. By identifying the key functions that estrogen controls, this research will reveal how estrogen promotes recovery and will guide the development of new treatments that mimic its benefits. Ultimately, these insights could lead to better, more personalized therapies for post-stroke recovery.
Jillian Saunders will explore how “magic mushrooms” provide rapid, long-lasting antidepressant effects, to develop better treatments for depression and related psychiatric disorders.
Image
PhD Candidate, Dept. of Neurobiology, Eroglu Laboratory
“Engagement of astrocytes by psilocybin to support rapid psilocybin-induced antidepressant effects”
Summary:
Psilocybin, or “magic mushroom,” is a psychedelic that has recently emerged as a potential treatment for depression.
A single dose of psilocybin has shown rapid, long-lasting antidepressant effects in clinical trials. This is promising, as current depression treatments have low efficacy, delayed effects, and require lifelong administration.
However, psilocybin is restricted by government regulations due to its hallucinogenic effects. Understanding how psilocybin produces therapeutic benefits is crucial for improving widespread access to this treatment. One reason for psilocybin’s therapeutic benefits may be related to how it interacts with astrocytes, the critical support cells that enable proper brain function. Interestingly, astrocytes are less active in the brains of depressed patients and have receptors that interact with psilocybin. I hypothesize that psilocybin activates astrocytes to achieve its rapid antidepressant effects. I will test this hypothesis using cells-in-a-dish and a mouse model of depression. These experiments will provide key insights into how psilocybin works and inform future therapeutic development for treating depression and related psychiatric disorders.
Edward Moseley will use computer modeling to create chemical tools to advance research on brain development, regeneration, and disease.
Image
PhD Candidate, Dept. of Computational Biology & Bioinformatics, Donald Lab
“Repurposing a bacterial motif as a selective probe of PKN2–Hippo signaling in neurons”
Summary:
The brain relies on carefully controlled signaling pathways to guide how neurons grow, repair themselves, and communicate. One of these pathways, called the Hippo pathway, helps regulate how nerve cells shape their structure, how stem cells in the brain divide, and how supporting cells respond after injury. Scientists know that a protein called PKN2 plays a key role in turning this pathway “off,” but there are currently no tools that allow researchers to adjust PKN2 activity in a precise and reversible way.
This project aims to create the first such tools. Using computer modeling, we will engineer a small piece of a bacterial protein that naturally interacts with PKN2. We will then test multiple versions of this protein in the lab to confirm that they work as intended.
By using computational design and biochemical experiments, we will create new chemical tools that can advance research on brain development, regeneration, and disease.