Early stages of drug discovery rely on docking programs to carry out computational screens to find novel potential binders for a protein target. Protein-ligand docking programs consist basically of a search algorithm and a scoring function. This chapter focuses on the search algorithm used in docking simulations. It explains how to use the differential evolution implemented in the Molegro Virtual Docker (MVD) to perform docking simulations against a protein target. Differential evolution is a heuristic method similar to genetic algorithms employed in optimization problems. This algorithm belongs to the class of bioinspired algorithms. This work describes a docking protocol that combines the differential evolution (a search algorithm) and the MolDock score (scoring function available in MVD). Previously published works reported docking simulations against three proteins (cyclin-dependent kinase 2, cannabinoid receptor 1, and transmembrane protease serine 2) using MVD. All docking simulations generated pose structures close to the crystallography coordinates of the ligands. Protein structures employed for docking in this chapter are available in the Protein Data Bank. The Jupyter Notebook discussed in this work is available at GitHub: https://github.com/azevedolab/docking#readme .