Abstract
OBJECTIVE: Despite decades of development in anti-seizure medications, ~30% of individuals remain refractory to all treatments, and none of the existing therapies are disease modifying. Identifying targets outside the current preclinical paradigm is critically important. This study aimed to characterize the landscape of current epilepsy treatments at the level of gene interaction networks and identify novel genetic modifiers of epilepsy as potential novel therapeutic targets. METHODS: We performed a functional network analysis to score genes based on their interactions with known epilepsy genes, and we integrated these functional scores with population genetics data and drug tractability information. In parallel, we performed a meta-analysis of genome-wide association studies of epilepsy-related phenotypes in genetically diverse mice using a large compendium of historical phenotyping data. Genes within mapped loci were prioritized based on functional rankings, and genomic evolutionary rate profiling (GERP) was used to identify highly single-nucleotide polymorphisms at evolutionarily constrained positions. RESULTS: Functional network analyses of known epilepsy genes revealed a strong involvement of neurodevelopmental processes in epilepsy pathogenesis, which are not targeted by existing or emerging treatments. Meta-analysis of seizure traits in mice identified 118 non-overlapping loci harboring potential seizure phenotype modifiers. Using functional rankings, we prioritized 168 candidate genes within these loci and used GERP scores to filter down to 75 SNPs as candidate variants within these genes. Among them, five genes-Ephb2, En2, Cadps2, Igsf21, and Cep170-contain regulatory variants in evolutionarily constrained sites. Four of these genes are validated as modifiers of neurological traits, including epilepsy susceptibility. SIGNIFICANCE: This study prioritized epilepsy modifier genes that are strongly predicted to influence neurodevelopmental processes, which are underrepresented among current therapeutic targets. Furthermore, the identified genes represent novel candidate modifiers with potential clinical relevance. Our systems-level analysis offers a novel view into the potential target landscape, pointing toward promising new directions for disease-modifying treatments.