Cyclin-dependent kinase 4 (CDK4) is a target for anticancer drugs, and available crystallographic structures made it possible to carry out docking simulations focused on this target. Also, the binding affinity data for CDK4 paved the way to build machine learning models to predict its inhibition. This chapter describes an integrated workflow to construct a neural network model to calculate the inhibition of CDK4 based on the atomic coordinates. This workflow utilizes the Molegro Data Modeller to build a regression model based on docking results of inhibitors for which binding affinity data is available (inhibition constant). The machine learning model relies on protein-pose structures determined using Molegro Virtual Docker (MVD). Ligands with experimental binding data are available from the BindingDB. Besides the pair of programs (MVD-MDM), this workflow discusses Jupyter Notebooks developed to integrate all steps necessary to build regression models to calculate binding affinity. All CDK4 datasets and Jupyter Notebooks discussed in this work are available at GitHub: https://github.com/azevedolab/docking#readme .