Elastic Net regression successfully builds computational models to address complex biological systems such as protein-drug complexes. Here, we explain the Elastic Net regression method and its application to model a protein system. Among the open-source libraries with Elastic Net, we focus our studies on the Scikit-Learn implementation. This library has tens of regression methods, including the Elastic Net. We examine the program SAnDReS 2.0, an open-source program designed to build regression models to predict enzyme inhibition and describe an Elastic Net regression model to calculate the inhibition of a protein target based on the atomic coordinates obtained through docking simulations. Also, we introduce the scoring function concept and how to implement the Elastic Net to explore it. We discuss a regression model to predict the inhibition of cyclin-dependent kinase 2. Our regression model shows superior predictive performance compared with a classical scoring function. All Jupyter Notebooks examined here are at GitHub: https://github.com/azevedolab/docking#readme . The program SAnDReS 2.0 is available at https://github.com/azevedolab/sandres .