32 Karge, Robin Tavosanis, Gaia Weber, Yvonne Gerda 165010 160000 535000-3 ; 933910 RWTH Aachen University Aachen English 1 Online-Ressource : Illustrationen Background: Epilepsies encompass a genetically and clinically heterogeneous group of disorders characterized by spontaneous recurrent seizures. While rare, monogenic variants account for many severe early-onset epilepsies, more common and less severe forms are largely shaped by polygenic influences. Despite advances in gene discovery, the functional validation of candidate genes and the clinical interpretation of polygenic risk remain limited. Bridging this gap requires integrative model systems and large-scale genomic data to support the development of precision medicine for epilepsy. Aim: This thesis aimed to investigate rare and common genetic mechanisms of epilepsy by leveraging functional modeling in Drosophila melanogaster and polygenic risk analyses in a large human epilepsy cohort. Methods: The monogenic epilepsy genes AP2M1 and STXBP1 were studied in Drosophila using pan-neuronal RNAi, CRISPR-based genome editing, and transgenic overexpression. Behavioral assays, electrophysiological recordings, and neuronal morphology were used to assess seizure susceptibility and synaptic function. For polygenic analyses, genome-wide SNP data from the Epi25 consortium was used to calculate polygenic risk scores across multiple traits and to identify genetically defined patient subgroups. Results & Conclusion: Loss of AP2M1 function resulted in a heat-induced paralysis phenotype and increased seizure resistance, resembling other defects in clathrin-mediated endocytosis in flies. The human pathogenic AP2M1 variant p.Arg170Trp only mildly altered seizure susceptibility without behavioral phenotypes. Elevated temperatures suppressed seizures but induced seizure-like discharges in AP-2μ-deficient flies. In the STXBP1 model, loss of the Drosophila ortholog Rop was lethal and not rescued by overexpression constructs, while STXBP1 overexpression partially rescued Rop-RNAi-induced lethality. In the human cohort, polygenic scores derived from non-epilepsy GWAS explained a portion of variance in epilepsy risk. Clustering based on multi-trait polygenic score profiles revealed distinct genetic subgroups with either cognitive or disease-related enrichment. Altogether, this work demonstrates how molecular modeling and genomic stratification can provide new insights into the complex genetic landscape of epilepsy and supports the integration of both approaches for future personalized treatment strategies. Veröffentlicht auf dem Publikationsserver der RWTH Aachen University ; Dissertation, RWTH Aachen University, 2025 ; 2, ; PUB:(DE-HGF)11, ; Dissertation / PhD Thesis Dissertation RWTH Aachen University 2025 2025 RWTH-2025-11108 2025 https://publications.rwth-aachen.de/record/1024277