Connection-Level Multi-Omics Reveals Cerebellar-Cortical Dysconnectivity and Divergent Molecular Signatures Across Major Psychiatric Disorders.
Abstract
Background
Understanding how disease-related connectome alterations relate to underlying molecular systems remains a major challenge in psychiatric disorders.
Methods
Here we introduce BrainNetAnno (https://brainnetanno.readthedocs.io/en/latest), an open-source framework for molecular annotation of brain network connections that extends multi-omics mapping from regional nodes to inter-regional edges. We applied this framework to resting-state functional connectivity data from a discovery cohort of 2,453 participants and an independent validation cohort of 442, including individuals with major depressive disorder, schizophrenia, generalized anxiety disorder, autism spectrum disorder, attention-deficit/hyperactivity disorder, and healthy controls.
Results
Our analyses revealed a shared transdiagnostic abnormal connectivity pattern (STACP), characterized by cerebellar-cortical dysconnectivity, alongside distinct disorder-specific connectivity deviations (DSCDs) that exhibited structured inter-disorder relationships, including opposition, partial alignment, and near-orthogonality. These connectivity patterns were differentially associated with molecular systems, with mitochondrial phenotypes linked to the STACP and selected DSCDs, and neurotransmitter receptor/transporter profiles primarily associated with DSCDs.
Conclusions
These findings establish a robust, connection-level framework for linking large-scale connectome abnormalities to multi-omics molecular systems in psychiatric disorders.