Rapid Screening of Anticoagulation Compounds for Biological Target-Associated Adverse Effects Using a Deep-Learning Framework in the Management of Atrial Fibrillation.

Results The evaluation of model performances shows that the SIDER dataset compounds showed a trend towards higher recall than precision ( Table 3 ), with rivaroxaban showing the highest recall (0.746), followed by dabigatran (0.702), and all models showing a higher precision and recall compared to the ConPlex model, which failed to attain satisfactory prediction in any of the tasks.

also in 53,322 other papers