Integrative bioinformatics and artificial intelligence analyses of transcriptomics data identified genes associated with major depressive disorders including NRG1.

Major depressive disorder (MDD) is a common mental disorder and is amongst the most prevalent psychiatric disorders. MDD remains challenging to diagnose and predict its onset due to its heterogeneous phenotype and complex etiology. Hence, early detection using diagnostic biomarkers is critical for rapid intervention. In this study, a mixture of AI and bioinformatics were used to mine transcriptomic data from publicly available datasets including 170 MDD patients and 121 healthy controls. Bioinformatics analysis using gene set enrichment analysis (GSEA) and machine learning (ML) algorithms were applied. The GSEA revealed that differentially expressed genes in MDD patients are mainly enriched in pathways related to immune response, inflammatory response, neurodegeneration pathways and cerebellar atrophy pathways. Feature selection methods and ML provided predicted models based on MDD-altered genes with ≥75% of accuracy. The integrative analysis between the bioinformatics and ML approaches identified ten key MDD-related biomarkers including NRG1, CEACAM8, CLEC12B, DEFA4, HP, LCN2, OLFM4, SERPING1, TCN1 and THBS1. Among them, NRG1, active in synaptic plasticity and neurotransmission, was the most robust and reliable to distinguish between MDD patients and healthy controls amongst independent external datasets consisting of a mixture of populations. Further evaluation using saliva samples from an independent cohort of MDD and healthy individuals confirmed the upregulation of NRG1 in patients with MDD compared to healthy controls. Functional mapping to the human brain regions showed NRG1 to have high expression in the main subcortical limbic brain regions implicated in depression. In conclusion, integrative bioinformatics and ML approaches identified putative non-invasive diagnostic MDD-related biomarkers panel for the onset of depression.

EEG Coherence Metrics for Vigilance: Sensitivity to Workload, Time-on-Task, and Individual Differences.

The vigilance decrement in performance is a significant operational issue in various applied settings. Psychophysiological methods for diagnostic monitoring of vigilance have focused on power spectral density measures from the electroencephalogram (EEG). This article addresses the diagnosticity of an alternative set of EEG measures, coherence between different electrode sites. Coherence metrics may index the functional connectivity between brain regions that supports sustained attention. Coherence was calculated for seven pre-defined brain networks. Workload and time-on-task factors primarily influenced alpha and theta coherence in anterior, central, and inter-hemispheric networks. Individual differences in coherence in inter-hemispheric, left intro-hemispheric and posterior networks correlated with performance. These findings demonstrate the potential applied utility of coherence metrics, although several methodological limitations and challenges must be overcome.

Use of EEG workload indices for diagnostic monitoring of vigilance decrement.

OBJECTIVE: A study was run to test which of five electroencephalographic (EEG) indices was most diagnostic of loss of vigilance at two levels of workload. BACKGROUND: EEG indices of alertness include conventional spectral power measures as well as indices combining measures from multiple frequency bands, such as the Task Load Index (TLI) and the Engagement Index (El). However, it is unclear which indices are optimal for early detection of loss of vigilance. METHOD: Ninety-two participants were assigned to one of two experimental conditions, cued (lower workload) and uncued (higher workload), and then performed a 40-min visual vigilance task. Performance on this task is believed to be limited by attentional resource availability. EEG was recorded continuously. Performance, subjective state, and workload were also assessed. RESULTS: The task showed a vigilance decrement in performance; cuing improved performance and reduced subjective workload. Lower-frequency alpha (8 to 10.9 Hz) and TLI were most sensitive to the task parameters. The magnitude of temporal change was larger for lower-frequency alpha. Surprisingly, higher TLI was associated with superior performance. Frontal theta and El were influenced by task workload only in the final period of work. Correlational data also suggested that the indices are distinct from one another. CONCLUSIONS: Lower-frequency alpha appears to be the optimal index for monitoring vigilance on the task used here, but further work is needed to test how diagnosticity of EEG indices varies with task demands. APPLICATION: Lower-frequency alpha may be used to diagnose loss of operator alertness on tasks requiring vigilance.