Edge-centric connectome representation reveal altered brain functional diversity in patients with major depressive disorder

Motivation: Traditional node-centric networks neglect edge-edge interactions, limiting the assessment of functional diversity patterns that may underlie critical neural mechanisms of major depressive disorder (MDD). Goal(s): Using edge-centric connectome model to investigate abnormal functional diversity in MDD and explore its neurobiological underpinnings and clinical significance. Approach: A cross-sectional, multi-site study of over 1600 participants, comparing regional functional diversity between MDD and controls. Neurobiological decoding and clinical contextualization analyses of abnormal functional diversity were conducted. Results: MDD patients showed increased functional diversity of widespread prefrontal and subcortical areas, which was neurobiologically related to neuroinflammation and serotonin. Edge-centric, compared with node-centric models exhibited better classification performance. Impact: This study demonstrated that edge-centric connectome could serve as a valuable complement to node-centric approaches, offering deeper insights into the neurobiological mechanisms related to functional diversity abnormalities in MDD and highlighting its promising clinical application value for diagnosis.