Automated reanalysis in rare disease

Project status: Active

Research area: Genomic Medicine > Translational Genomics

Iterative re-analysis of genomic data from undiagnosed patients with rare disease, at scale

Reanalysis of genomic data in rare disease is a powerful way to find diagnoses for families affected by rare disease. Currently, this is an intensive, manual process meaning that few families benefit.

Automation has the potential to scale regular reanalysis to thousands of patients and make it routine, meaning all patients benefit equally from advances in knowledge and improvements in data analysis methods.

Automation has the potential to scale regular...

The challenges

Genomic testing has transformed rare disease diagnosis, enabling more families to receive timely and accurate answers to guide care. However, over 50 percent of individuals remain undiagnosed after initial testing.

Unlike most diagnostic tests, genomic data can be stored and reanalysed over time, with strong evidence that this leads to new diagnoses. Despite policy support, reanalysis remains limited in practice due to complex workflows, workforce constraints, and lack of reimbursement. As a result, only a small proportion of patients benefit, with access often inequitable.

Automation offers a pathway to scale reanalysis across thousands of datasets and deliver it more consistently. However, key challenges remain around implementation at scale and balancing accuracy, workload, frequency and level of automation.

Study details

We conducted a landscape analysis to understand current attitudes and practice in Australia surrounding reanalysis. This included focus groups, audit of laboratory data, and a workforce survey.

In partnership with the Centre for Population Genomics and other national and international collaborators we designed and evaluated an automated tool, called Talos, that performs iterative reanalysis of rare disease datasets, identifying new diagnoses as knowledge about genetic conditions and analysis methods improve.

About Talos

Talos is a scalable, open-source variant prioritisation tool designed to automate reanalysis of genomic data for rare disease. It periodically re-examines stored data from previously undiagnosed patients, enabling new insights over time.

Used worldwide, Talos has reanalysed data from more than 10,000 patients (including 5,000 from Australia) and contributed to over 350 diagnoses, with adoption spanning the USA, UK, Denmark, Belgium, Germany and Hong Kong.

By continually revisiting existing data, it helps deliver new answers and hope to families.

Funders & collaborators

Key collaborators on this project include:

Project datasets were generated by:

Australian Genomics
Rare Genomes Project
Victorian Clinical Genetics Services (VCGS)

The project was funded by the Australian Government’s Genomics Health Futures Mission (MRF2008820).

Research team

Professor Zornitza Stark, Clinical Geneticist at the Victorian Clinical Genetics Services (VCGS) and Co-Lead of the Translational Genomics research group, MCRI
Professor Daniel MacArthur, Director of the Centre for Population Genomics
Dr Cas Simons, Rare Disease Analysis Lead at the Centre for Population Genomics
Associate Professor Sebastian Lunke, Genetics & Genomics Innovation, VCGS and Co-Lead of the Translational Genomics research group, MCRI
Dr Simon Sadedin, Head of Clinical Bioinformatics, Bioinformatics Group, MCRI