Comprehensive Summary
Migraines affect nearly one billion people worldwide. Along with migraines comes a high comorbidity of psychiatric disorders such as major depressive disorder (MDD). The objective of this research was to investigate the “molecular mechanisms linking migraine and MDD and to prioritize shared blood-based biomarkers.” In order to achieve this, the team deployed two machine learning algorithms, LASSO (Least Absolute Shrinkage and Selection Operator) regression and Support Vector Machine-Recursive Feature Elimination (SVM-RFE). They were able to identify Pentraxin 3 (PTX3) and haptoglobin (HP) as hub genes that could serve as diagnostic biomarkers. The accuracy of the biomarkers was validated using Receiver Operating Characteristic (ROC) curves which further demonstrated the usefulness in using PTX3 and HP to distinguish the control group from the MDD group. Specifically, PTX3 showed high accuracy in identifying migraines with an AUC of 0.912.
Outcomes and Implications
The identification of PTX3 and HP as substantial biomarkers can allow for earlier identification of the comorbid migraine and MDD. Once again, we see the application of machine learning being put to use in the identification of biomarkers. Each breakthrough can bring us closer to more personalized treatment plans by identifying niche biomarkers.