Pancreatic cancer is recognized as the fourth leading cause of death. Treatment cost for pancreatic cancer remains very high. So, searching for novel plant compounds with potent anticancer activity is required. Computer-aided drug design has gained importance recently. Using these techniques drugs for specific targets can be predicted in silico. In this present study, PARP from pancreatic cancer was chosen as the target, and plant alkaloids from the literature were selected as ligands. First, the alkaloid was screened for their drug-likeness and pharmacokinetics properties, and violations of Lipinski's rule were rejected. Docking studies were carried out to analyze the compounds with the best binding affinity, and the best compound was selected for analyzing toxicity profiles. Compounds with toxic endpoints were rejected and final lead compounds were identified; the identified leads were again screened for bioavailability and molecular target predictions. The results are compared with control drugs. All compounds showed drug-likeness and pharmacokinetics except, Geissospermine. From docking studies, 15 compounds showed the best binding affinity of -6.0 to -8.8 Kcal/mol. Atropine, ephedrine, theobromine, theophylline, actinidine, and pinidine have no toxic endpoints. These compounds were predicted as final lead compounds. Leads also possesses oral bioavailability, which was predicted by radar plot. The identified hits were analyzed for molecular targets. Atropine, ephedrine, theobromine, theophylline, actinidine, and pinidine were predicted as hit among 21 compounds but further in vitro and in vivo studies are required to validate its action.