Understanding the genetic basis of clinical heterogeneity in Parkinson’s disease

Parkinson’s disease is notoriously variable. In the clinic, patients experience markedly different ages at onset, symptom manifestations (both motor and non-motor), comorbidities, and treatment responses. This sheer clinical diversity is a major bottleneck—it complicates prognosis and hinders the development of therapies that work effectively for everyone.

Thanks to massive global genetic studies, the field has a solid grasp of the variants associated with the risk of developing Parkinson’s. However, we know surprisingly little about the genetic architecture that drives how the disease presents and evolves once a patient is diagnosed. Do specific genetic variants dictate rapid clinical decline or severe adverse reactions to medication? Furthermore, do these genetic effects hold true across different global populations? Historically, research has heavily skewed toward populations of European descent, leaving a critical gap in our understanding.

This project tackles that gap using the Australian Parkinson’s Genetics Study (APGS). We pair deep genomic data with highly detailed, participant-reported clinical and lifestyle phenotypes. Because our group is closely integrated with the Global Parkinson’s Genetics Program (GP2), you will be driving analyses within a genuinely international, ancestrally diverse framework. You will also have the logistical backing to validate your findings across the world's most powerful complementary cohorts, including the UK Biobank, PPMI, the Million Veteran Program, and other datasets.

Our primary goal is to map the genetic factors driving clinical diversity in Parkinson’s disease and ultimately determine how these expression factors relate to the baseline genetic architecture of disease risk.

Potential research questions include:

1.    Which specific genetic variants drive age at onset, distinct symptom profiles, and treatment responses?

2.    Do clinically recognised subtypes of Parkinson’s actually possess distinct genetic blueprints?

3.    How do cumulative polygenic influences shape a patient's clinical presentation, and can we use polygenic scores to stratify meaningful patient subgroups?

4.    Which genetic effects are shared globally, and which are specific to certain ancestry groups?

Note: The student won't be expected to tackle all of this alone. We will map out your precise thesis questions together, tailoring them to your unique skills, interests, and the maturity of the available phenotype data.

Day-to-day, you will be working hands-on with massive genomic and phenotypic datasets from APGS and GP2. A major early component of your research will involve the logistical challenge of defining, harmonising, and validating clinically meaningful phenotypes across these different international cohorts.

Analytically, you will deploy a wide toolkit. This ranges from foundational statistical genetics (GWAS, polygenic scoring,
genetic correlation) to advanced approaches like genomic structural equation modelling, causal inference, and multi-trait analyses. Where scientifically appropriate and adequately powered, we also encourage the integration of machine learning and clustering methods.

Throughout the project, you will build robust expertise in high-performance computing (HPC) and Linux-based environments, programming primarily in R and/or Python. Beyond the code, you will refine your scientific writing and learn how to navigate and collaborate within large-scale national and international consortia.

The big-picture objective here is to move the field of Parkinson's research beyond a simple case–control definition. By identifying the genetics behind how the disease presents, your findings could
unearth new biological pathways, dramatically improve the design of future genetic studies, and lay the groundwork for patient stratification in precision clinical trials. We design every thesis stream around clearly defined, impactful questions, so you can expect to produce strong, peer-reviewed publications as a central outcome of your PhD.

Ideal Candidate

This project is an excellent fit for a driven student with a background in data science, computer science, statistics, genetics, biomedical science, or epidemiology. While prior programming experience is highly desirable, what we value most is a strong quantitative aptitude, a logical approach to problem-solving, and a genuine commitment to mastering computational skills.