 |
 |
Consistently coupled QM/MM calculations started with Warshel and Levitt 1976 work that considered an entire enzyme-substrate complex in solution and evaluated the catalytic effect of this system. Our subsequent studies paved the way for quantitative modeling of enzymatic reactions and chemical reactions in solutions. These works have developed and refined hybrid quantum/molecular mechanics (QM/MM) methods that covered the many options of multiscale embedding strategies. This included the EVB with its powerful ability to obtain reliable free energy surfaces, the ab initio QM/MM (QM(ai)/MM) that used the EVB as a reference potential in free energy calculations and the frozen DFT (FDFT) and constraint DFT (CDFT) that allow us to represent all the protein on the ab initio level3.
While large protein structures found in living things are composed of many thousands of molecules, only a few places in a protein are usually chemically important. The quantum mechanics/molecular mechanics (QM/MM) approach combines classical simulation (MM) for the large structural parts of the protein with quantum simulation (QM) for the very small subset of molecules that are chemically active. The system energy is the sum of the three energies:
E = E_{MM} + E_{QM} + E_{QM/MM}
where E_{MM} is the energy within the classical region, E_{QM} is the energy within the quantum region, and E_{QM/MM} is a classical electrostatic interaction energy between the two regions. Different ways of treating the coupling between the QM and MM regions have been proposed. In 1976, the seminal work by Warshel and Levitt introduced a consistent coupling for the study of enzymic reactions in solvent, [1] [2] enabling multiscale modeling for large biological molecules over decades of increasing computational power. Our group continuously works on refining and validating QM/MM strategies for use in many of the proteins that we study [3].
|
|