Hepatitis C is a disease that constitutes a serious global health problem, is often asymptomatic and difficult to diagnose, about 60-80% of infected patients develop chronic diseases over time. As there is no vaccine against hepatitis C virus (HCV), developing new cheap treatments is a big challenge. The search for new drugs from natural products has been outstanding in recent years. The aim of this study was combining structure-based and ligand-based virtual screening (VS) techniques to select potentially active molecules against two HCV target proteins from in-house secondary metabolite dataset (SistematX). From the ChEMBL database, we selected two sets of 1199 and 237chemical structures with inhibitory activity against different targets of HCV to create random forest models with an accuracy value higher than 72% for cross-validation and test sets. Afterward, a ligand-based virtual screen of the entire 1848 secondary metabolites database stored in SistematX was performed. In addition, a structure-based virtual screening was also performed for the same set of secondary metabolites using molecular docking. Finally, using consensus analyzes approach combining ligand-based and structure-based VS, two alkaloids and one triterpene were selected as potential anti-HCV compounds.