Hepatitis B virus (HBV) infects more than 400 million humans Worldwide. Currently, development of new anti-HBV agents is focused on inhibiting of HBV DNA polymerase activity. The natural components of medicinal plant have a broad spectrum of biological activities with therapeutic properties which can be exploited in various steps of drug discovery. Currently, in silico analyses have been introduced as alternative or supplements methods for drug discovery. This study was planned to in silico screening novel HBV DNA polymerase inhibitor(s) from R. palmatum, R. coreanus and S. officinalis. For this purpose, a set of dominant phytochemicals from mentioned plants were retrieved from PubChem database and primary screening was performed with molecular docking method using iGemdock 2.1 software. SwissADME and MedChem Designer 3.0 were used to calculate the drug-likeness parameters of the ligands. Furthermore, the genotoxicity of the studied ligands was predicted using Toxtree 2.6.6 software. Final analysis of screened compounds was done using Autodock 4 software. Result confirmed that Frangulosid and Lindleyin acid have most and least efficacy in HBV DNA polymerase inhibition with the inhibition constant of 2.97 and 53.83 µM, respectively. Results also showed that, the amino acids, involved in interaction, were different for each compound. In this regards, results revealed that the main amino acids residues of the receptor, involved in interaction with Quercetin-3-glucuronide, Frangulosid and Lindleyin separately, located in 420–424, 606–615 and 512–542 spectra, respectively. In conclusion, Frangulosid can be considered as a good candidate for more investigation of its anti-HBV activity.