Bioinformatics Seminar
Time: 11AM
Venue: Davis Auditorium and Online
3 March 2026
This is a WEHI only event.Improving Diagnosis, Treatment and Understanding of Paediatric Epilepsy using AI-Based Integrative Omics
Ehsan RamezaniWEHI
Epilepsy is one of the most common neurological disorders, affecting millions of children worldwide. Despite major global efforts, two critical challenges remain: low diagnostic yield and limited understanding of the underlying disease mechanisms. In this work, we address both challenges using patient-derived samples obtained through clinical testing pathways and epilepsy surgery. First, we developed and evaluated a new computational diagnostic pipeline designed to improve diagnostic yield in undiagnosed patients of paediatric epilepsy. Our results demonstrate the feasibility and clinical utility of integrating emerging experimental approaches into diagnostic workflows, with evidence of meaningful impact in previously unresolved cases. Second, we investigated disease mechanisms in surgically resected epileptogenic tissue from patients with focal cortical dysplasia (FCD), a leading cause of drug-resistant epilepsy in children, using single-nucleus and spatial transcriptomic profiling. By applying state-of-the-art AI-based data integration and rare-cell discovery strategies, we focused on identifying disease-relevant cellular populations within highly heterogeneous brain tissue. We then used spatial transcriptomics and histological annotation to validate and contextualise these observations within the tissue microenvironment. Together, this work highlights two complementary advances in paediatric epilepsy research: improved computational approaches for increasing diagnostic yield, and AI-enabled multi-modal molecular profiling and integration for investigating disease mechanisms in drug-resistant epilepsy. Our results demonstrate the value of combining advanced computational methods with patient data to improve disease characterisation and support future therapeutic development.