Quantifying Pleiotropy through Directed Signaling Networks: A Synchronous Boolean Network Approach and In-Silico Pleiotropic Scoring.

Abstract

Pleiotropy refers to a gene's ability to influence multiple phenotypes or traits. In the context of human genetic diseases, pleiotropy manifests as different pathological effects resulting from mutations in the same gene. This phenomenon plays a crucial role in understanding gene-gene interactions in system-level biological diseases. Previous studies have largely focused on pleiotropy within undirected molecular correlation networks, leaving a gap in examining pleiotropy induced by directed signaling networks, which can better explain dynamic gene-gene interactions. In this study, we utilized a synchronous Boolean network model to explore pleiotropic dynamics induced by various mutations in large-scale networks. We introduced an in-silico Pleiotropic Score (sPS) to quantify the impact of gene mutations and validated the model against observational pleiotropy data from the Human Phenotype Ontology (HPO). Our results indicate a significant correlation between sPS and network structural characteristics, including degree centrality and feedback loop involvement. The highest correlation was observed between closeness centrality and sPS (0.6), suggesting that genes more central in the network exhibit higher pleiotropic potential. Furthermore, genes involved in feedback loops demonstrated higher sPS values (p < 0.0001), supporting the role of feedback loops in amplifying pleiotropic behavior. Our model provides a novel approach for quantifying pleiotropy through directed network dynamics, complementing traditional observational methods.