AutoDock4 and AutoDockTools4 are automated docking software tools that allow for the prediction of biomolecular complexes. AutoDock4 incorporates limited flexibility in the receptor, enabling more accurate modeling of protein-ligand interactions. The software was tested using 188 diverse ligand-protein complexes and 87 HIV protease complexes, demonstrating its effectiveness in predicting binding conformations and energies. It also provides a method for analyzing covalently-bound ligands using grid-based docking and a modified flexible sidechain technique. AutoDock4 uses a Lamarckian genetic algorithm for conformational searching, allowing for the exploration of a large conformational space. It incorporates a semiempirical free energy force field to predict binding free energies of small molecules to macromolecular targets. The software allows for fully flexible modeling of specific portions of the protein, enabling the simulation of torsional degrees of freedom. This flexibility is particularly useful for analyzing covalently-attached ligands. AutoDockTools is a graphical user interface that facilitates the preparation, design, and analysis of docking experiments. It provides a range of methods for clustering, displaying, and analyzing docking results. AutoDockTools is implemented in Python and is built from reusable software components, making it accessible for a wide range of applications, including automated high-throughput screening. The validation of AutoDock4 was performed using two sets of complexes: 188 diverse protein-ligand complexes and 87 HIV protease complexes. The results showed that AutoDock4 successfully redocks most complexes with about 10 or fewer torsional degrees of freedom, but fails for most complexes with higher conformational flexibility. In cross docking experiments, flexible docking improved docking in complexes that were most expected to benefit. However, the results also showed that flexible sidechains increased the size of the conformational space to be searched, leading to more false positives. The results of covalent docking experiments showed that a grid-based approach and a flexible sidechain method were both effective in predicting the binding of covalently-attached ligands. The flexible sidechain method provided excellent results, with all 10 docking runs giving similar conformations that were very similar to the crystallographic conformation. Overall, AutoDock4 and AutoDockTools4 provide a powerful tool for automated docking and analysis of biomolecular complexes.AutoDock4 and AutoDockTools4 are automated docking software tools that allow for the prediction of biomolecular complexes. AutoDock4 incorporates limited flexibility in the receptor, enabling more accurate modeling of protein-ligand interactions. The software was tested using 188 diverse ligand-protein complexes and 87 HIV protease complexes, demonstrating its effectiveness in predicting binding conformations and energies. It also provides a method for analyzing covalently-bound ligands using grid-based docking and a modified flexible sidechain technique. AutoDock4 uses a Lamarckian genetic algorithm for conformational searching, allowing for the exploration of a large conformational space. It incorporates a semiempirical free energy force field to predict binding free energies of small molecules to macromolecular targets. The software allows for fully flexible modeling of specific portions of the protein, enabling the simulation of torsional degrees of freedom. This flexibility is particularly useful for analyzing covalently-attached ligands. AutoDockTools is a graphical user interface that facilitates the preparation, design, and analysis of docking experiments. It provides a range of methods for clustering, displaying, and analyzing docking results. AutoDockTools is implemented in Python and is built from reusable software components, making it accessible for a wide range of applications, including automated high-throughput screening. The validation of AutoDock4 was performed using two sets of complexes: 188 diverse protein-ligand complexes and 87 HIV protease complexes. The results showed that AutoDock4 successfully redocks most complexes with about 10 or fewer torsional degrees of freedom, but fails for most complexes with higher conformational flexibility. In cross docking experiments, flexible docking improved docking in complexes that were most expected to benefit. However, the results also showed that flexible sidechains increased the size of the conformational space to be searched, leading to more false positives. The results of covalent docking experiments showed that a grid-based approach and a flexible sidechain method were both effective in predicting the binding of covalently-attached ligands. The flexible sidechain method provided excellent results, with all 10 docking runs giving similar conformations that were very similar to the crystallographic conformation. Overall, AutoDock4 and AutoDockTools4 provide a powerful tool for automated docking and analysis of biomolecular complexes.