What is molecular docking?

Molecular docking is an important method for structure-based drug design and screening by studying the interaction of ligand and receptor molecules and predicting their affinity and binding modes.

Elements in molecular docking (Hernández et al., 2013)

Why molecular docking?

Identification of protein-protein interactions (PPIs) is of great importance for revealing important signaling pathways, studying protein functions, discovering new targets for drug action, and designing drugs targeting PPIs.

There are already many experimental methods to study PPIs, such as immunoprecipitation, yeast two-hybrid, and bimolecular fluorescence complementation. However, it is difficult to help us reveal protein-protein interaction information at the atomic level by relying only on biophysical or biochemical experimental techniques.

The crystal structure of protein-protein complexes can provide the most accurate information about protein-protein interactions. Several crystallographic experimental techniques can be used for crystal structure resolution, including X-ray crystallography, nuclear magnetic resonance (NMR) spectroscopy, and cryo-EM techniques. However, most proteins, especially protein-protein complexes, still do not have corresponding 3D structures in the Protein Structure Data Bank (PDB).

In addition, many weak or transient PPIs play a large role in the dynamic regulatory network of biological systems. Due to the instability of their complex structure, they are difficult to be captured by experimental means. Therefore, protein-protein docking, as an important adjunct to experimental techniques, is by far the most important technique.

In the field of drug screening, proteomics and molecular docking techniques can be used in combination. Proteomics helps us to find the differential proteins between samples and the corresponding functions of the differential proteins. Molecular docking helps to understand the interactions between molecules.

The key proteins that are directly related to many differential proteins can be found by protein-protein interaction network (PPI) maps. PPI is only a prediction and does not really indicate that there are interactions between proteins, so further validation is needed. Using molecular docking, it is possible to determine in advance whether key proteins and proteins have corresponding binding sites.

Principle of molecular docking

Ligand-receptor binding must satisfy the principle of mutual matching, i.e., ligand and receptor geometry, electrostatic, hydrogen bonding, and hydrophobic interactions are complementarily matched.

Molecular docking process

Molecular docking places molecules from a database of known 3D structures one by one at the active site of the target molecule. By continuously optimizing the position of the receptor compound, its conformation, the dihedral angle of the rotatable bonds within the molecule and the side chains and backbone of the amino acid residues of the receptor, the best conformation for the receptor small molecule compound to interact with the target macromolecule is searched and its binding mode and affinity are predicted. The ligands with the best affinity to the receptor are selected by scoring functions that are close to the natural conformation.

Outline of the molecular docking process (Ferreira et al., 2015).

(A) Three-dimensional structure of the ligand; (B) Three-dimensional structure of the receptor; (C) The ligand is docked into the binding cavity of the receptor and the putative conformations are explored; (D) The most likely binding conformation and the corresponding intermolecular interactions are identified.

Types of molecular docking

(1) Rigid docking: The rigid docking method does not change the conformation of the molecules involved in docking during the calculation, and only changes the spatial position and attitude of the molecules. The rigid docking method has the highest degree of simplicity and relatively small computational effort, and is suitable for dealing with docking between large molecules.

(2) Semi-flexible docking: The semi-flexible docking method allows a certain degree of change in the conformation of small molecules during docking, but usually fixes the conformation of large molecules. In addition, the adjustment of the small molecule conformation may be limited to some extent, such as fixing the bond lengths and bond angles of some non-critical parts. The semi-flexible docking method is one of the more widely used docking methods as it balances the computational volume with the predictive power of the model.

(3) Flexible docking: The flexible docking method allows the conformation of the studied system to change freely during the docking process. Since the variables grow geometrically with the atomic number of the system, the flexible docking method is very computationally intensive and suitable for accurate examination of intermolecular recognition.