A diagnostic model combining five serum proteins (HECTD3, NID2, FTO, GOLM1, and NPL) effectively distinguished patients with vascular depression from controls with an AUC of 0.9046.
Case-Control (n=71)
No
A novel 5-protein serum biomarker panel identified via machine learning shows high accuracy in distinguishing patients with vascular depression from non-depressed controls.
Background: Vascular depression (VaD) is a depressive disorder closely associated with cerebrovascular disease and vascular risk factors. It remains underestimated owing to challenging diagnostics and limited information regarding the pathophysiological mechanisms of VaD. The purpose of this study was to analyze the proteomic signatures and identify the potential biomarkers with diagnostic significance in VaD. Methods: Deep profiling of the serum proteome of 35 patients with VaD and 36 controls was performed using liquid chromatography-tandem mass spectrometry (LC-MS/MS). Functional enrichment analysis of the quantified proteins was based on Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, and Reactome databases. Machine learning algorithms were used to screen candidate proteins and develop a protein-based model to effectively distinguish patients with VaD. Results: ≤ 0.05). Enrichment pathways analyses showed that neurobiological processes related to synaptic vesicle cycle and axon guidance may be dysregulated in VaD. Extrinsic component of synaptic vesicle membrane was the most enriched term in the cellular components (CC) terms. 19 candidate proteins were filtered for further modeling. A nomogram was developed with the combination of HECT domain E3 ubiquitin protein ligase 3 (HECTD3), Nidogen-2 (NID2), FTO alpha-ketoglutarate-dependent dioxygenase (FTO), Golgi membrane protein 1 (GOLM1), and N-acetylneuraminate lyase (NPL), which could be used to predict VaD risk with favorable efficacy. Conclusion: This study offers a comprehensive and integrated view of serum proteomics and contributes to a valuable proteomics-based diagnostic model for VaD.
Lan et al. (Wed,) conducted a case-control in Vascular depression (n=71). 5-protein diagnostic model (HECTD3, NID2, FTO, GOLM1, NPL) vs. Non-depressive controls was evaluated on Diagnostic accuracy (AUC) in the training cohort (95% CI 0.8257-0.9834). A diagnostic model combining five serum proteins (HECTD3, NID2, FTO, GOLM1, and NPL) effectively distinguished patients with vascular depression from controls with an AUC of 0.9046.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: