Abstracts

Rationale:
Omics technologies enable an unprecedented view of gene expression events associated with disease progression and treatment, including epilepsy. A challenge with these technologies is interpretation of lengthy lists of differentially expressed (DE) genes in the appropriate biological context. For neurological studies, there are currently no available computational tools that analyze DE genes to identify their impact on neuronal electrophysiology and associated gene networks and pathways. We created a bioinformatics tool (Enkefalos) to address this critical gap. Using a published dataset of DE genes in epilepsy of tuberous sclerosis complex (TSC), we demonstrate how Enkefalos can illuminate altered genes and networks that are involved in neuronal electrophysiology.

Methods:
Enkefalos was developed using Python programming and will be publicly available through GitHub. Enkefalos takes as input a list of DE genes and queries published datasets to identify input genes that have a statistically significant correlation between gene expression and electrophysiological properties. Statistical significance and false discovery rate (FDR) are user selectable. Identified genes are analyzed for cellular pathways and networks using the STRING interaction database. The number of network interactions are tabulated for each gene to identify key nodes, or potential cellular drivers, identified among the input list.

Results:
We used Enkefalos to analyze a large list of DE genes from a previous study of human brain tissue associated with epilepsy in TSC patients. We first selected 3977 genes that were significantly over-expressed in epileptogenic TSC tissue. This list was input into Enkefalos which then identified 1235 genes having significant correlations with neuronal electrophysiological properties. Pathway and network analysis of the 1235 genes revealed several key nodes involved in EGFR signaling that may contribute to altered neuronal electrophysiology in TSC epilepsy.

Conclusions:
Omics analyses are important for advancing our understanding of cellular events that contribute to neurological disorders. However, there is currently a lack of bioinformatics tools to interpret lists of DE genes in the context of neuronal electrophysiology. We have developed a bioinformatics tool that fills this void. To demonstrate the utility of Enkefalos we used the tool to identify genes implicated in epilepsy of TSC that also have significant correlations to electrophysiological properties. Key nodes among the genes were identified and revealed that several genes in the EGFR pathway may play a key role in TSC seizures. We anticipate that Enkefalos will be widely used in a broad range of neurological studies to provide new biological insights and hypotheses.

Funding:
This research was supported by grant W81XWH-22-1-0798 of the Department of Defense TSC CDMRP Program.