AUTHOR=Gong Yuandong , Lu Zhe , Kang Zhewei , Feng Xiaoyang , Zhang Yuyanan , Sun Yaoyao , Chen Weimin , Xun Guanglei , Yue Weihua 

TITLE=Peripheral non-enzymatic antioxidants as biomarkers for mood disorders: Evidence from a machine learning prediction model

JOURNAL=Frontiers in Psychiatry

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2022.1019618

DOI=10.3389/fpsyt.2022.1019618

ISSN=1664-0640

ABSTRACT=Background: Oxidative stress is related to the pathogenesis of mood disorders, and the level of oxidative stress may differ between bipolar disorder (BD) and major depressive disorder (MDD) may be different. This study aimed to detect the differences in non-enzymatic antioxidants levels between BD and MDD, and assessed the predictive value of non-enzymatic antioxidants in the prediction of mood disorders by applying a machine learning model. 
Methods: Peripheral uric acid (UA), albumin (ALB), and total bilirubin (TBIL) of were measured in 1188 participants (discover cohort: 157 with BD and 544 with MDD; validation cohort: 119 with BD and 95 with MDD; 273 healthy controls) were measured. An Extreme Gradient Boosting (XGBoost) model and a logistic regression model were used to assess the predictive effect. 
Results: All 3 indices differed between patients with mood disorders and healthy controls; additionally, the UA levels of BD were higher than those of MDD. After treatment, UA levels increased in the MDD group, while it decreased in the BD group. Finally, we entered age, sex, UA, ALB and TBIL into the XGBoost model. The area under the curve (AUC) of the XGBoost model for distinguishing between BD and MDD reached 0.849 (Accuracy = 0.808, 95% CI = 0.719–0.878), and for distinguishing between bipolar disorder with depression episode (BD-D) and MDD was 0.899 (Accuracy = 0.891, 95% CI = 0.856–0.919). The models were validated in the validation cohort. The most important feature distinguishing between BD and MDD was UA. 
Conclusion: Peripheral non-enzymatic antioxidants, especially the UA, might be a potential biomarker capable of distinguishing between BD and MDD.