Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Aug 8:16:949609.
doi: 10.3389/fnins.2022.949609. eCollection 2022.

A machine learning model for predicting patients with major depressive disorder: A study based on transcriptomic data

Sitong Liu  1   2 , Tong Lu  3 , Qian Zhao  1   2 , Bingbing Fu  1   2 , Han Wang  1   2 , Ginhong Li  1   2 , Fan Yang  1   2 , Juan Huang  1   2 , Nan Lyu  1   2
Affiliations

A machine learning model for predicting patients with major depressive disorder: A study based on transcriptomic data

Sitong Liu et al. Front Neurosci. .

Abstract

Background: Identifying new biomarkers of major depressive disorder (MDD) would be of great significance for its early diagnosis and treatment. Herein, we constructed a diagnostic model of MDD using machine learning methods.

Methods: The GSE98793 and GSE19738 datasets were obtained from the Gene Expression Omnibus database, and the limma R package was used to analyze differentially expressed genes (DEGs) in MDD patients. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed to identify potential molecular functions and pathways. A protein-protein interaction network (PPI) was constructed, and hub genes were predicted. Random forest (RF) and artificial neural network (ANN) machine-learning algorithms were used to select variables and construct a robust diagnostic model.

Results: A total of 721 DEGs were identified in peripheral blood samples of patients with MDD. GO and KEGG analyses revealed that the DEGs were mainly enriched in cytokines, defense responses to viruses, responses to biotic stimuli, immune effector processes, responses to external biotic stimuli, and immune systems. A PPI network was constructed, and CytoHubba plugins were used to screen hub genes. Furthermore, a robust diagnostic model was established using a RF and ANN algorithm with an area under the curve of 0.757 for the training model and 0.685 for the test cohort.

Conclusion: We analyzed potential driver genes in patients with MDD and built a potential diagnostic model as an adjunct tool to assist psychiatrists in the clinical diagnosis and treatment of MDD.

Keywords: artificial neural network; bioinformatics analysis; machine learning; major depressive disorder; random forest.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
Before and after box diagram of standardization of GSE19738 and GSE98793 datasets. (A) The flow chart of this study. (B) Box diagram of the GSE19738 dataset before correction; (C) Box diagram of the GSE98793 dataset before correction; (D) Box diagram of the GSE19738 dataset after correction; and (E) Box diagram of the GSE98793 dataset after correction; Red represents the MDD samples, and blue represents the normal samples.
FIGURE 2
FIGURE 2
DEG identification of the GSE19738 dataset. (A) DEGs of peripheral blood samples from patients with MDD and healthy controls were obtained from the DEG heat map constructed from the GSE19738 dataset. Horizontal coordinate blue represents the control group, red represents the experimental group, blue indicates low expression, and red indicates high expression. (B) Volcano diagram, black indicates genes with no differential expression, blue indicates down-regulated genes, and red indicates up-regulated genes. (C) GO enrichment analysis. The outer circle represents the number of GO term, the outer circle number represents all genes in GO term, and the inner circle number represents the number of enriched genes. The inner circle pie chart represents the percentage of genes that are enriched. (D) KEGG pathways. The outer circle represents the KEGG ID, the outer circle number represents all genes in the KEGG pathway, and the inner circle number represents the number of genes enriched in the pathway. The inner circle pie chart represents the percentage of genes that are enriched. DEGs, differentially expressed genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.
FIGURE 3
FIGURE 3
(A) Protein-protein interaction network of DEGs constructed using cytoscape. (B) The top 10 hub genes were explored using CytoHubba.
FIGURE 4
FIGURE 4
(A) Random forest tree. (B) MeanDecreaseGini. (C) Artificial Neural Network model. Healthy stands for healthy group, and MDD stands for MDD group.
FIGURE 5
FIGURE 5
Receiver operating characteristic curves for the artificial neural network. The AUC curve of GSE19738 training cohort is on the left, and the AUC curve of GSE98793 testing cohort is on the right.
FIGURE 6
FIGURE 6
Evaluation and correlation analysis of immune cell infiltration. (A) Panoramic view of 22 immune cell infiltrates in peripheral blood samples; (B, C) High and low expression group of immune cell infiltration difference.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Arbore G., West E., Spolski R., Robertson A., Klos A., Rheinheimer C., et al. (2016). T helper 1 immunity requires complement-driven NLRP3 inflammasome activity in CD4+ T cells. Science 352:aad1210. 10.1126/science.aad1210 - DOI - PMC - PubMed
    1. Bhak Y., Jeong H., Cho Y., Jeon S., Cho J., Gim J., et al. (2019). Depression and suicide risk prediction models using blood-derived multi-omics data. Transl. Psychiatry 9:262. 10.1038/s41398-019-0595-2 - DOI - PMC - PubMed
    1. Bouchon A., Facchetti F., Weigand M., Colonna M. (2001). TREM-1 amplifies inflammation and is a crucial mediator of septic shock. Nature 410 1103–1107. 10.1038/35074114 - DOI - PubMed
    1. Ciobanu L. G., Sachdev P. S., Trollor J. N., Reppermund S., Thalamuthu A., Mather K. A., et al. (2020). Downregulated transferrin receptor in the blood predicts recurrent MDD in the elderly cohort: a fuzzy forests approach. J. Affect Disord. 267 42–48. 10.1016/j.jad.2020.02.001 - DOI - PubMed
    1. DiSabato D., Nemeth D., Liu X., Witcher K., O’Neil S., Oliver B., et al. (2021). Interleukin-1 receptor on hippocampal neurons drives social withdrawal and cognitive deficits after chronic social stress. Mol. Psychiatry 26 4770–4782. 10.1038/s41380-020-0788-3 - DOI - PMC - PubMed

LinkOut - more resources