Using Retinal Imaging and Genetics to Reveal Early Markers of Parkinson’s Disease

Why the retina? A window into the brain

The retina is part of the central nervous system, offering a unique, non-invasive way to observe neural tissue. Previous research has shown that people with Parkinson’s often have thinning in specific retinal layers.

However, it has been unclear whether these changes occur before symptoms, and whether they reflect underlying genetic risk.

The team investigated this by combining retinal imaging with powerful genomic analysis using Setonix supercomputer.

Their aim was to determine whether genetic susceptibility to Parkinson’s is associated with changes in retinal structure, and whether these eye-based measures could serve as early biomarkers of neurodegeneration.

A uniquely Australian dataset

The study used data from the Raine Study, a long-running West Australian cohort that has tracked thousands of participants from pregnancy into adulthood.

Researchers analysed retinal scans taken at ages 20 and 28 using optical coherence tomography (OCT), which provides precise measurements of retinal layers such as the ganglion cell inner plexiform layer (GCIPL) and the retinal nerve fibre layer (RNFL).

Because these participants are young adults, decades before the typical onset of Parkinson’s, the dataset provides a rare opportunity to investigate early structural changes long before symptoms could appear.

The study also incorporated findings from the world’s largest Parkinson’s disease genome-wide association study, which includes genetic information from more than 1.4 million individuals.

What the researchers found

The study revealed that individuals with higher genetic risk for Parkinson’s showed greater thinning of retinal layers between ages 20 and 28, particularly in the GCIPL.

This suggests that structural changes in the retina may begin far earlier in life than previously thought.

The team also identified shared genetic regions influencing both Parkinson’s disease risk and retinal structure, pointing toward common biological pathways that could be targeted in future research.

These findings strengthen the case for using the eye as an early detection tool for neurodegenerative disease. They also highlight the value of population-based cohorts and comprehensive genetic studies working together to unlock new health insights.

Why high-performance computing was essential

The work required complex, large-scale computational workflows that cannot run efficiently on standard servers or commercial cloud platforms. This included calculating polygenic risk scores from millions of genetic variants, performing genome-wide colocalisation analyses, modelling genetic correlations, and integrating multi-omics datasets.

The Pawsey Supercomputing Research Centre provided the secure, high-memory, high-throughput computing environment necessary to conduct these analyses. With fast parallel processing, powerful nodes, and optimised scientific software environments, Pawsey allowed the team to process multi-terabyte datasets and computationally intensive statistical models with speed and reliability.

Just as importantly, using a sovereign HPC facility ensured that Australian health and genetic data remained within national borders. For research involving human genetics, long-term cohort studies, and Indigenous data considerations, onshore processing is essential for maintaining ethical, legal, and community trust.

A strong case for Australian HPC

Research like this increasingly relies on the ability to combine global scientific datasets with nationally health information.

By providing sovereign high-performance computing infrastructure, Australia enables researchers to analyse data locally while collaborating internationally. This approach supports world-leading biomedical research while maintaining control over data governance, privacy, and security.

For fields such as genomics, medical imaging, and precision health, where datasets are growing rapidly in size and complexity, national supercomputing capabilities are becoming an essential part of the research ecosystem.

Looking ahead

The study opens new possibilities for detecting Parkinson’s disease much earlier in life, potentially allowing interventions before significant neurological damage occurs.

As retinal imaging technologies advance and genomic datasets continue to expand, research of this kind will push the limits of data analysis even further.

With advanced systems such as the Setonix supercomputer, Australian researchers have the tools needed to explore these complex datasets and accelerate discoveries in neurodegeneration, precision medicine, and early disease detection.