Abstract 6
Category: Basic Science
At the end of the session, participants will be able to:
- Apply deep learning to nominate candidate regions of interest for spatial transcriptomics analysis
- Identify molecular drivers of excessive excitatory signalling in cellular drivers of focal cortical dysplasia
- Produce a publicly available comprehensive transcriptional resource of key pathologic features in focal cortical dysplasia
COI Disclosure:
None to disclose.
Presenter
Ameesha Paliwal is a graduate student in the Laboratory Medicine and Pathobiology program at the University of Toronto. Under the mentorship of Dr. Phedias Diamandis in the Princess Margaret Cancer Centre, Ameesha’s research focuses on using deep learning to model tissue heterogeneity in neuropathology disease systems, including focal epilepsy. In her work, she demonstrates that by accounting for heterogeneity, researchers are able to scrutinize cellspecific drivers of poor outcomes and support discovery of precision medicine-based therapeutic approaches.
Authors
Ameesha Paliwal1,2, Kevin Faust2, Okty A. Borhani2, Parsa B. Zadeh2, Phedias Diamandis1,2,3,4
1 Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON M5S 1A8, Canada.
2 Princess Margaret Cancer Centre, 101 College Street, Toronto, ON M5G 1L7, Canada.
3 Department of Medical Biophysics, University of Toronto, 101 College St, Toronto, ON M5G 1L7, Canada
4 Laboratory Medicine Program, Department of Pathology, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada.
Target Audience:
Pathologists, Residents, Medical Students
CanMEDS:
Communicator
An anatomic transcriptional atlas of focal cortical dysplasia
Abstract
Focal cortical dysplasia (FCD) is a severe neurodevelopmental malformation that manifests in medically refractory epilepsy. FCD presents with classic histological hallmarks, such as cytomegalic dysmorphic neurons, that are presumed to be key anatomical drivers of disease. The relatively low frequency and interspersed patterns of dysmorphic neurons within resected lesions has challenged traditional molecular characterization approaches and consequently, therapeutic options. Therefore, the objective of this study is to systematically identify the key anatomical components of FCD in multiple tissue sections in order to perform molecular sequencing on dysmorphic neuron-enriched samples. To accomplish this, we leveraged unsupervised deep learning approaches to map the conventional pathological features of this disease for spatially-conserved RNA-sequencing. Using this approach, we present the highest resolution molecular catalog of FCD to date. Our analysis has uncovered non-canonical signaling and neurotransmitter pathways in dysmorphic neurons that could serve as new targets for this debilitating disorder. Specifically, we have found aberrant enrichment of an inhibitory glycinergic neurotransmission regulator, glycine transporter 2 (GlyT2). Immunohistochemical staining in eight additional cases of FCD revealed precise enrichment for GlyT2 in dysmorphic neurons. GlyT2 inhibitors have demonstrated efficacy in treating central nervous system disorders by reducing excessive excitatory signaling, making this transporter a compelling therapeutic approach for further research in FCD. Importantly, we have generated a comprehensive resource for discovery in FCD, which reflects the key pathologic features of this disease including cytomegalic dysmorphic neurons, balloon cells, and rarefacted white matter. We anticipate that the research and resource produced here will improve outcomes in FCD.