The purpose of this study is to examine if aqueous autotaxin (ATX) and TGF-β levels could be used for differentiating glaucoma subtypes. This prospective observational study was performed using aqueous humor samples obtained from 281 consecutive patients. Open angle glaucoma patients were classified into three groups: primary open-angle glaucoma (POAG), secondary open-angle glaucoma (SOAG), and exfoliation glaucoma (XFG). Aqueous levels of ATX and TGF-βs were quantified. The AUC as well as sensitivity and specificity for the classification into normal and glaucoma subtypes using four indicators-ATX, TGF-β1, TGF-β2, and TGF-β3, upon the application of three machine learning methods. ATX, TGF-β1, and TGF-β3 were positively correlated with IOP, and ATX was significantly and negatively correlated with the mean deviation. From least absolute shrinkage and selection operator regression analysis, the AUC values to distinguish each subgroup [normal, POAG, SOAG, and XFG] ranged between 0.675 (POAG vs. normal) and 0.966 (XFG vs. normal), when four variables were used. High AUC values were obtained with ATX for discriminating XFG from normal eyes and with TGF-β3 for discriminating XFG from normal eyes, POAG, or SOAG. Aqueous TGF-β and ATX exhibited high diagnostic performance in detecting glaucoma subtypes, and could be promising biomarkers for glaucoma.