Brain network-driven predictive neural markers for suicide risk
- 1Experimental Therapeutics and Pathophysiology Branch, National Institute of Mental Health, National Institutes of Health
- 2Center for Depression Research and Clinical Care, Peter O’Donnell Jr. Brain Institute and Department of Psychiatry, University of Texas Southwestern Medical Center
- 3Magnetoencephalography Core Facility, National Institute of Mental Health, National Institutes of Health
Abstract
Background: Suicide is a significant public health challenge, underscoring the need for temporally precise and objective biomarkers to identify acute suicide risk. One potential biomarker is electrophysiological brain connectivity measured by magnetoencephalography (MEG). This study used publicly available atlases to estimate parcel-level within-network connectivity.
Methods: The multi-omics statistical method was applied to assess whether canonical networks—including the default mode (DMN), executive control (ECN), and salience (SN) networks—could differentiate individuals experiencing a recent suicidal crisis (high risk) from those with a history of suicide attempts but not in the past year (lower risk), clinical controls with depression but no history of suicide risk (clinical controls), and individuals with no personal or family history of either psychiatric disorder or suicidal behavior (minimal risk).
Results: The high risk group exhibited significantly weaker functional connectivity within specific regions of these networks ( ps <.05). Further analyses revealed that recent suicidal ideation scores were positively associated with functional DMN connectivity (e.g., between the precuneus/posterior cingulate cortex and dorsal/medial prefrontal cortex) and SN connectivity (e.g., between the medial frontal cortex and medial parietal cortex and between the parietal operculum and lateral prefrontal cortex) ( ps <.05). Conversely, functional ECN connectivity (between the medial/posterior prefrontal cortex and precuneus) was negatively associated with recent suicidal ideation ( ps <.05).
Conclusions: These findings highlight critical brain regions and network interactions, advancing our understanding of the neural mechanisms underlying suicide risk. Leveraging electrophysiological connectivity and innovative tools may help develop targeted interventions and early identification strategies for those at heightened risk of suicide.
Clinical trials registration: www.clinicaltrials.gov , identifier NCT02543983.
Related
- pyhgf: A neural network library for predictive coding
- Functional and Structural Brain Age Gaps as Complementary Predictive Markers for Cognitive Performance
- Regional and network-level resting state functional connectivity markers of pharmacological treatment resistance in obsessive-compulsive disorder
- A predictive model based on maternal immune-inflammatory markers for ARDS identification in mechanically ventilated late-preterm and term neonates
- Data-driven predictive design of engineered living hydrogels
BibTeX
@article{lee2026brain,
title = {{Brain network-driven predictive neural markers for suicide risk}},
author = {Yoojin Lee and Jessica R. Gilbert and Cherise R. Chin Fatt and Jeffrey D. Stout and Laura R. Waldman and Steven J. Lamontagne and Carlos A. Zarate and Elizabeth D. Ballard},
journal = {Frontiers in Psychiatry},
year = {2026},
month = oct,
doi = {10.3389/fpsyt.2026.1950068},
url = {https://www.frontiersin.org/articles/10.3389/fpsyt.2026.1950068/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Brain network-driven predictive neural markers for suicide risk
AU - Yoojin Lee
AU - Jessica R. Gilbert
AU - Cherise R. Chin Fatt
AU - Jeffrey D. Stout
AU - Laura R. Waldman
AU - Steven J. Lamontagne
AU - Carlos A. Zarate
AU - Elizabeth D. Ballard
PY - 2026
DA - 2026/10/20
JO - Frontiers in Psychiatry
DO - 10.3389/fpsyt.2026.1950068
UR - https://www.frontiersin.org/articles/10.3389/fpsyt.2026.1950068/full
AB - Background: Suicide is a significant public health challenge, underscoring the need for temporally precise and objective biomarkers to identify acute suicide risk. One potential biomarker is electrophysiological brain connectivity measured by magnetoencephalography (MEG). This study used publicly available atlases to estimate parcel-level within-network connectivity. Methods: The multi-omics statistical method was applied to assess whether canonical networks—including the default mode (DMN), executive control (ECN), and salience (SN) networks—could differentiate individuals experiencing a recent suicidal crisis (high risk) from those with a history of suicide attempts but not in the past year (lower risk), clinical controls with depression but no history of suicide risk (clinical controls), and individuals with no personal or family history of either psychiatric disorder or suicidal behavior (minimal risk). Results: The high risk group exhibited significantly weaker functional connectivity within specific regions of these networks ( ps <.05). Further analyses revealed that recent suicidal ideation scores were positively associated with functional DMN connectivity (e.g., between the precuneus/posterior cingulate cortex and dorsal/medial prefrontal cortex) and SN connectivity (e.g., between the medial frontal cortex and medial parietal cortex and between the parietal operculum and lateral prefrontal cortex) ( ps <.05). Conversely, functional ECN connectivity (between the medial/posterior prefrontal cortex and precuneus) was negatively associated with recent suicidal ideation ( ps <.05). Conclusions: These findings highlight critical brain regions and network interactions, advancing our understanding of the neural mechanisms underlying suicide risk. Leveraging electrophysiological connectivity and innovative tools may help develop targeted interventions and early identification strategies for those at heightened risk of suicide. Clinical trials registration: www.clinicaltrials.gov , identifier NCT02543983.
SN - 1664-0640
ER -