San Diego’s high-powered computers are doing more than crunching climate models and galaxy maps. A new brain simulation study suggests that a single type of cortical cell may be the primary cause of the beta-frequency brain waves that afflict people with Parkinson’s disease, disrupting their movement.
Researchers report that when one specific neuron type, known as PT5B cells, goes off script, those troublesome beta oscillations spike. By zeroing in on PT5B dysfunction, the team believes they have identified a far more precise cellular target than today’s treatments, which mostly replace dopamine or broadly stimulate brain regions and networks.
As reported by UC San Diego Today, scientists in the Aligning Science Across Parkinson’s (ASAP) Collaborative Research Network pushed massive virtual brains through the Expanse supercomputer at the San Diego Supercomputer Center using NSF ACCESS allocations. Their simulations suggest that reduced PT5B activity creates a traffic jam in the motor system, choking off normal communication. “To create a faithful model of Parkinson’s disease in the brain, we started with data from rodent models and fed it into sophisticated simulations powered by Expanse,” said Donald W. Doherty, a research scientist on the team. The work hits close to home: the researchers estimate Parkinson’s costs California more than $6 billion a year in healthcare and lost productivity.
Detailed Models Tie PT5B Cells To Parkinson’s Beta Waves
The peer-reviewed study, published in npj Parkinson’s Disease, used NEURON/NetPyNE simulations to test what happens when PT5B neurons start underperforming. The models showed that relatively small decreases in PT5B neuron excitability can drive an order-of-magnitude jump in beta-band power around 15 Hz in simulated primary motor cortex.
That surge in beta oscillations, a hallmark of Parkinson’s motor problems, appeared even when only a limited subset of neurons was altered, the authors report. The paper argues that PT5B cells occupy a critical junction in the motor system, serving as a final common pathway to downstream motor circuits. That role, the team says, makes them a compelling cellular bullseye for future therapies.
Supercomputing Power Turned Brain Theory Into Testable Code
Those detailed, biologically grounded simulations needed serious hardware. Expanse provides researchers nationwide with mixed CPU and GPU resources through the NSF ACCESS program, according to the San Diego Supercomputer Center’s documentation. The system is designed to handle heterogeneous, data-intensive workloads and can support both GPU-accelerated model training and large-scale neuronal simulations without slowing down.
UC San Diego Today notes that the Parkinson’s team ran its Expanse jobs using an ACCESS allocation (No. IBM140002), which covered the cost of the enormous compute hours needed to track cell-level dynamics in the virtual cortex.
What This Could Mean For Future Parkinson’s Care
Current Parkinson’s therapies tend to flood the system with dopamine substitutes or rely on broad-brush stimulation of brain regions, strategies that can help but also bring side effects. A cell-specific intervention targeting PT5B neurons could, in theory, refine that approach, allowing patients to achieve better mobility with fewer unwanted trade-offs, according to the researchers.
Parkinson’s Foundation estimates that more than 1.1 million Americans are living with Parkinson’s disease, which highlights how widely any new, more precise treatment approach could ripple if it proves effective. Adaptive deep brain stimulation and other closed-loop neuromodulation techniques are already beginning to tune stimulation based on brain signals, so PT5B-focused strategies could eventually integrate with those platforms as they mature.
The Aligning Science Across Parkinson’s network states that it plans to transition these simulation results into experimental and clinical testing, aiming to “restore normal neuron function or compensate for their loss,” according to the ASAP summary, as reported by UC San Diego Today. With the San Diego Supercomputer Center and UC San Diego labs providing both the raw computational power and cross-disciplinary expertise, the region is well-positioned to carry this work from code to clinic. If animal and human studies confirm the models, future patients may see treatments that repair disrupted circuitry at the cellular level, rather than just masking symptoms on the surface.
