Machine learning methods have great potential to build models to address protein-ligand interactions obtained through docking simulations. Molegro Data Modeller (MDM) has an intuitive interface with Molegro Virtual Docker (MVD), which allows us to integrate docking results and machine learning. Here, we present a tutorial on how to build a regression model using a support vector machine to predict the inhibition of cyclin-dependent kinase 2 with MDM. Our model relies on docked poses of CDK2 inhibitors obtained with MVD and employs the support vector machine implemented in the MDM program. We focus on ligands for which binding affinity data is available at the BindingDB and the structure of a CDK2-Cyclin complex determined using crystallography. Our approach explores the concept of scoring function space to build targeted models. We take descriptors, energy terms, and scoring functions determined with MVD to build our machine learning model. All CDK2 datasets and Jupyter Notebooks discussed in this work are available at GitHub: https://github.com/azevedolab/docking#readme .