Alterations in edge-centric functional connectivity in patients with major depressive disorder and their genetic...

Previous studies on brain functional networks in Major Depressive Disorder (MDD) have mainly focused on node changes, but the dynamics of edge-to-edge connectivity remain unclear. This study combines edge-centric functional connectivity (eFC) and whole-brain transcriptomics to reveal higher-order network interactions in MDD.

We enrolled 163 MDD patients and 135 healthy controls (HCs). First, time series were extracted to construct the functional connectivity (FC) matrix. Then, edges were extracted from this matrix, and their Pearson correlation coefficients were calculated to construct the eFC matrix. Between-group differences in eFC were compared. Subsequently, support vector machines (SVM), random forest (RF) and extreme gradient boosting (XGBoost) models were built to evaluate the classification performance of eFC in diagnosing MDD. Finally, by integrating transcriptomic data, we identified genes whose spatial expression profiles were associated with eFC alterations and performed functional enrichment analysis.

We observed that compared to HCs, there are extensive changes in eFC. Specifically, individuals with MDD exhibited increased eFC in the left Superior frontal gyrus, right Middle frontal gyrus and bilateral Inferior temporal gyrus, while displaying decreased eFC in the bilateral Caudate nucleus. The classification results demonstrated that models based on eFC features outperformed those based on traditional FC in key metrics, and this advantage remained stable across different algorithms. Partial least squares (PLS) analysis revealed that alterations in eFC in MDD patients are associated with specific gene expression profiles. These genes were significantly enriched in pathways related to ion channels and synaptic transmission. These findings were replicated in validation cohort and HarvardOxford brain atlas.

Our study revealed alterations in the eFC network in MDD patients and their associations with gene expression profiles, providing a novel perspective to advance the understanding of MDD.