A Novel Machine Learning Method and Drug for the Diagnosis and Treatment of Depression
ISEF · 2025 Translational Medical Science
Overview
Major Depressive Disorder negatively affects millions worldwide by decreasing productivity, degrading physical well-being, and increasing the risk of suicide by 60%. Moreover, miscalculating the risk of suicide after rehabilitation, by inaccurately diagnosing depression, accounts for many suicide deaths. Studies find that the gene expression levels significantly affect the development of depression. However, current antidepressants increase serotonin and dopamine levels in the brain, not addressing the root cause of depression. This study provides a statistical analysis using machine learning to diagnose depression and identify a target gene to produce a drug that combats it. A Convolutional Neural Network (CNN) was used to identify patterns between an NIH RNA-sequencing dataset and depression labels. Next, Principal Component Analysis (PCA) was used to categorize the various differentially expressed genes (DEGs) by their influence on depression. To increase the model’s accuracy, SMOTE was used to augment samples of the minority class for the training data, while original samples were used for testing. Using these methods, a Convolutional Neural Network (CNN) model achieved an accuracy of 93.24%. Next, PCA identified the top genes contributing to a depression diagnosis by assessing differential expression, identifying MT-ND4 as the top contributor. Protein transduction pathway analysis revealed that inhibiting 7UDQ activated the production of ND4, the protein made by MT-ND4, mitigating depression. The chemical composition of a drug that inactivates 7UDQ with a lower binding energy (-8.5 kcal/mol) than the original ligand (-8.2 kcal/mol) resulted from molecular docking simulations of several experimental ligands.
Competition history
- ISEF 2025
Resources
Related projects
ISEF · 2026
A Multi-Omics and Machine Learning Approach for Identifying Potential Salivary Biomarkers of and Treating Major Depressive Disorder
ISEF · 2019
A Novel Noninvasive and Inexpensive Biomarker for Diagnosing Major Depressive Disorder (MDD): Using Machine Learning Model in silico and Drosophila melanogaster Model in vivo
ISEF · 2020
Exploring the Mechanisms of Major Depression and Antidepressant Response Using Gene and miRNA Expression
CWSF · 2026
Machine Learning Prediction of SSRI Response Using fMRI Connectivity
Closest projects by meaning, across every fair and year in the corpus.
Source: Regeneron International Science and Engineering Fair