|
|
Research Interests
My research is focussed on studying chemical reactions in solution
and in enzymes using molecular dynamics simulation techniques with combined QM/MM methodology.
The reactions include phosphate hydrolysis reactions
in water, in hairpin ribozyme, and in ATP synthase, as well as other chemical reactions.
The phosphate hydrolysis is one of the most important reactions in biological system,
but relatively unknown in their actual chemical mechanisms.
The requirement of extensive sampling of configuration space using molecular dynamics limits the
use of high level ab initio electronic structure method in QM/MM potential.
This limts our choice of quantum methods on semiempirical models for QM/MM calculations.
The main difficulty in studying this reaction is from the lack of accuracy
in currently available semiempirical models to study phosphate hydrolysis and hydrogen bonding.
The MNDO/d Hamiltonian is relatively reliable in phosphate hydrolysis reaction.
However, this model has well-known difficulty in
handling hydrogen bonding and proton transfer reactions that are closely connected with
many phosphate hydrolysis reactions of interests.
In addition, the reactions in RNA and many highly charged systems require rigorous
evaluation of their long-range electrostatic interactions.
The Ewald summation method is one way to calculate the long range electrostatic
interactions correctly in periodic boundary conditions.
However, most used combined QM/MM methods do not include this long range interactions
extensively.
Currently, my specific focuses are on three ways,
Developement of Ewald Sum Method in Combined QM/MM Potential
This project is to include the Ewald sum technique into QM/MM method.
This requires several approximations, such as monopolar interactions of electronic density with
Ewald sum potential generated from all the image atoms in entire crystal. This is reasonably simple
and reliable approximation, since interactions between two electron densities at large separation
approach Coulomb interactions between two point charges.
Development of New Semiempirical method and Tests
The inclusion of d orbital in MNDO/d semiempirical Hamiltonian makes it relatively reliable in
the di-anionic transphosphorylation reaction mechanism.
Our group has new semiempirical models such as AM1/d and PM3/d semiempirical Hamiltonians,
which will be better in hydrogen bonded systems and/or hydrogen transfer reactions.
In one part, I am wokring on developing Reaction Specific Parameters (SRP) of phosphorous atom
in phosphate hydrolysis reaction using MNDO/d and AM1/d Hamiltonians by using ab initio database that
other group members have been generated using density functional theory.
In second part, the New Semiempirical method will have changes
in the basic NDDO approximations and the long-range dispersion effects, as well as full parameterizations
on their parameters against high level density functional calculations.
In addition, this semiempirical model will be tested with combined QM/MM simulations.
Development of Multi-layered QM/MM Method
We are currently faced with fast increases in computing resources and their speed, which will make it
possible the direct ab initio QM/MM simulations in the future though it is still too time comsuming.
However, the difficulty in obtaining reliable semiempirical models limits the type of systems that can be studied.
The multi-layered QM/MM method can overcome these two limitations.
This method use ab initio method for relatively small region in the system, use semiempirical method for
relatively large region, and molecular mechanical method for the rest of system.
The main idea of multi-layered method is similar to ONIOM method developed from Morokuma group.
The difference in the method is in their handling inter-regional interactions.
Our method use combined QM/MM scheme, in which high level region interacts with MM or lower level region
directly. Thus, it is more realistic in including the effects from environment than ONIOM method.
 |
"Dynamics of an enzymatic substitution reaction in
haloalkane dehalogenase", Kwangho Nam, Xavier Prat-Resina, Mireia Garcia-Viloca,
Lakshmi S. Devi-Kesavan, and Jiali Gao, J. Am. Chem. Soc.,
126, 1369-1376, (2004).
|
|
"Solvent and protein effects on the vibrational frequency shift and
energy relaxation of the azide ligand in carbonic anhydrase",
Mireia Garcia-Vilova, Kwangho Nam, Cristobal Alhambra, and Jiali Gao, J. Phys. Chem. B,
108, 13501-13512, (2004).
|
|
"A Combined QM/MM approach to protein-ligand interactions:
Polarization effects of the HIV-1 protease on selected high affinity inhibitors",
Christian Hensen, Johannes C. Hermann, Kwangho Nam, Shuhua Ma, Jiali Gao, and
Hans-Dieter Höltje, J. Med. Chem.,
47, 6673-6680, (2004). |
|
"An efficient linear-scaling Ewald method for long-range
electrostatics in combined QM/MM calculations",
Kwangho Nam, Jiali Gao, and Darrin M. York, J. Chem. Theory. Comput.,
1, 2-13, (2005). |