Dual inhibition of A(2A) and MAO-B is an emerging strategy in neurodegenerative diseases, such as Alzheimer's disease (AD) and Parkinson's disease (PD). In this study, atom-based three-dimensional quantitative structure-activity relationship (3D-QSAR) and hologram quantitative structure-activity relationship (HQSAR) models were generated with benzothiazine and deazaxanthine derivatives. Based on activity against A(2A) and MAO-B, two statistically signi?cant 3D-QSAR models (r(2) = 0.96, q(2) = 0.76 and r(2) = 0.91, q(2) = 0.63) and HQSAR models (r(2) = 0.93, q(2) = 0.68 and r(2) = 0.97, q(2) = 0.58) were developed. In an activity cliff analysis, structural outliers were identified by calculating the Mahalanobis distance for a pair of compounds with A(2A) and MAO-B inhibitory activities. The generated 3D-QSAR and HQSAR models, activity cliff analysis, molecular docking and dynamic studies for dual target protein inhibitors provide key structural scaffolds that serve as building blocks in designing drug-like molecules for neurodegenerative diseases.