Abstract
Understanding how the brain's functional connections develop during infancy is crucial for uncovering the complexities of early neural maturation. Traditional node-based analyses have advanced our knowledge, but may overlook the transient dynamics of interregional connectivity. Leveraging the large neonatal functional MRI dataset from the Developing Human Connectome Project (n = 781, including 494 full-term and 287 preterm infants), we introduce an edge-centric metric to quantify cross-module functional integration. Here we show that preterm infants exhibit higher edge participation coefficients than full-term peers, suggesting delayed network specialization. We mapped developmental changes in edge participation coefficients and found that between-network connections-particularly those involving visual and higher-order systems-undergo the most pronounced changes and are associated with cognitive outcomes at 18 months. By analyzing gene expression in a developing brain, we identified genes involved in neurodevelopmental processes and cellular signalling that may underlie these patterns. Our findings illustrate how interregional diversity evolves in early life and provide insight into the molecular basis of early brain development.