Docking (chemistry)

Molecular docking ( docking for short, dt .: coupling, fitting) is a bioinformatics / chemoinformatics method with which the binding mode and ideally also the binding energy of two biomolecules that bind to one another is predicted. Docking is typically used in molecular biology and pharmaceutical research. A main area of ​​application is the search for drug candidates for a pharmaceutically relevant problem. With the help of docking methods, large amounts of substance can be tested virtually for binding to a target molecule. These in silico bonding studies are significantly faster and more cost-effective than comparable methods in the wet laboratory, but usually of noticeably lower accuracy.

The prediction takes place on the basis of the already known chemical and spatial structure of the two starting molecules. The structure and the released binding energy of the complex that is formed from the two molecules is unknown before the calculation. As a solution, you get the best possible approximation of the complex structure and, depending on the method, an estimate of the binding energy of the complex.

The molecules involved are usually highly complex, which is why an a priori calculation of the bound state is currently not possible due to the enormous computing capacities required for this. Therefore, all relevant algorithms use strong to very strong approximations of the underlying physics or chemistry in order to calculate a good estimate of the binding mode.

There are different approaches to keeping the complexity of the problem in check. The simplest structural approach is rigid docking, which is based on the simplifying assumption that the molecules involved do not change spatially during the binding process. There are also approaches that keep one of the binding partners rigid while the second is (partially) flexibly modeled. The latter is often used in ligand docking, since here small, active ingredient-like molecules are considered, the comparatively low spatial complexity of which makes flexible consideration manageable.

Since we are dealing with three-dimensional objects, the mutual positioning and orientation of the molecules in the bound state, ie in the state of lowest energy, is of interest. In addition, the molecules themselves can experience changes in their conformation during the process . The parameters of this unknown final state could also be determined experimentally, but this is too complex; therefore, computer-aided methods are used to calculate the virtual space of the biochemical complex and to come as close as possible to real conditions.

Docking is divided into sub-areas depending on the type of binding partner

• Protein-protein docking,
• Protein ligand docking and
• Docking of DNA or RNA.

This subdivision is necessary because the properties of the binding partners involved in the complex require the use of special algorithms .