Synergistic Coactivation Probabilities of Large-Scale Resting-State Networks in Major Depressive Disorder.

Area of Science:

• Neuroscience
• Computational Psychiatry
• Systems Neuroscience

Background:

• Major depressive disorder (MDD) presents with complex, distributed alterations in large-scale brain networks.
• Understanding synergistic network patterns is crucial for identifying MDD drivers.
• Integrative approaches are needed to analyze complex brain network interactions.

Purpose of the Study:

• To investigate simultaneous resting-state network (RSN) activation patterns, or co-activation, in individuals with and without MDD.
• To explore the relationship between dynamic network organization and clinical symptoms in MDD.
• To apply a dynamical systems approach to resting-state fMRI data.

Main Methods:

• Utilized a pairwise Maximum Entropy Model to estimate RSN co-activation probabilities from resting-state fMRI data.
• Analyzed data from 867 participants: 487 healthy controls (HC), 175 with current MDD (cMDD), and 205 with remitted MDD (rMDD).
• Examined co-activation patterns across seven RSNs, including the default mode network (DMN), frontoparietal network (FPN), and visual network (VIS), capturing 128 possible network states.

Main Results:

• Elevated co-activation probabilities were observed in cMDD, particularly involving the DMN, FPN, and VIS.
• A specific co-activation state (DMN, VIS, dorsal attention network [DAN], FPN, language network [LAN]) strongly correlated with MDD diagnosis and symptom severity.
• Canonical Correlation Analysis identified two network-symptom profiles linking DMN/DAN co-activation to cognitive/mood symptoms and sensorimotor network (SMN)/VIS co-activation to cognitive/somatic symptoms.

Conclusions:

• MDD, especially acute episodes, is characterized by a dominance of DMN, FPN, and VIS co-activation, indicating altered dynamic network organization.
• Changes in brain state dynamics are significantly associated with MDD symptoms.
• This study underscores the importance of dynamic network analysis in understanding MDD pathophysiology.