A computational analysis using graph theory has mapped key metabolic and behavioral interactions underlying GDM. The study, published in Diabetology, integrated data from 44 clinically relevant factors identified through a PRISMA-guided literature review spanning 2004–2025. A directed network was constructed in Python to model associations among these variables and to quantify their systemic roles using centrality metrics and Minimum Dominating Set (MDS) analysis.

The MDS comprised 20 structurally dominant nodes, with vitamin D and sedentary lifestyle emerging as the most influential. Vitamin D exhibited 15 outgoing links, connecting positively with protective elements such as high-density lipoprotein and negatively with risk factors including obesity and smoking. Closeness centrality emphasized triglycerides, insulin resistance, uric acid, and fasting plasma glucose as strongly predictive nodes within the network.

The findings highlight vitamin D status and lifestyle behavior as major control points within the GDM framework. This graph-based approach provides a systems-level perspective for developing early predictive tools and targeted preventive interventions in maternal metabolic health.