Early and accurate prediction of adverse drug reactions (ADRs) by computer-aided methods is extremely important for decreasing health threats, along with preventing product withdrawal after clinical trials which need long time and high cost. In this study, we built a model to predict 329 known ADRs of 27 approved antidepressant drugs, and we examined three machine learning algorithms (multilayer perceptron, k-nearest neighbour and support vector machine) in order to find which one is better for this duty. We integrated the known ADRs of a drug with the drug's chemical structures and biological properties, including drug targets, enzymes and transporters for prediction of ADR in our predictive model. In addition, 329 known ADRs were grouped according to their system organ classes (SOCs) and compared the performances of ADR predictions. This assessment, based on a ten-fold cross-validation, showed that the multilayer perceptron (MLP) algorithm surpassed the others. 248 out of 329 ADRs were successfully predicted. The best performing attribute group was the chemical plus biological, followed by the chemical and biological. The mean AUC values are 0.623, 0.676, and 0.695 for the feature groups of chemical, biological, and chemical plus biological, respectively. Among ADRs of the 21 SOCs, only ADRs belonging to 5 SOCs were predicted successfully by chemical, biological or combined attributes. Also, the ADRs related to withdrawal of indalpine, pheniprazine, medifoxamine, zimelidine and amineptine were successfully predicted by our model constructed using MLP with the CfsSubsetEval and BestFirst, based on chemical properties of the approved antidepressants. The results showed the external validity of our approach, predicting satisfactory amount of previously known ADRs from the literature. In conclusion, the proposed approach is an efficient and promising tool for predicting ADRs of antidepressants.