|
Our group's research is dedicated to understanding, analysing, and more importantly, quantitatively predicting the structure, energy and interaction dynamics of complex chemical and biological systems at the microscopic molecular level using theoretical/computational methods developed from
first principles.
Although tremendous progress has been achieved by experimentalists in chemistry
and biology, the advance in theoretical and computational chemistry is
still relatively slow in comparison. This is due to the enoumous
mathematical and numerical difficities resulting from the complexitiy of
chemical and biological systems. In recent years, our group
has made significant advance in developing quantitative computational methods
to predict structures and dynamics of small molecular systems in gas phase
and gas surface. Our present research goal is to significantly extend
computational capabilities to more complex chemical and biological systems without loosing the necessary accuracy.
|
MFCC Method for Full Ab Initio Computation of Protein-Ligand Interaction Energy
We recently developed an efficient MFCC (molecular fractionation with conjugate caps) method to compute protein-ligand, DNA-ligand or other large molecular interaction energies fully ab initio. The MFCC method decomposes a protein molecule
into amino acid-based fragments that are properly capped. As a result, the
interaction energy between a protein and a ligand can be efficiently obtained by
separate computation of fragment-ligand interaction energies and subtracted by
conjugate caps-ligand interaction energies. This method enables one to carry
out practical full ab initio quantum chemistry calculation of protein-ligand
interaction energies at desired levels of theory (HF, DFT, MP2, or higher).
The MFCC calculation scales linearly with the size of the protein molecules
and can be trivially parallelized. Using the MFCC method, practical
protein-ligand systems with thousands of atoms can now be computed routinely on
standard workstations. In addition, the MFCC calculation provides a natural means to gain molecular insight into the chemical nature of protein-ligand binding by providing a quantitative analysis of individual fragment-ligand interaction energies. This type of quantitative analysis is extremely useful in rational design of drugs.
Related publications:
MFCC method:
Protein-ligand interaction:
Quantum mechanical computation of protein energy:
Quantum mechanical computation of protein-drug binding :
MFCC-CPCM method for protein solvation:
A new quantum method for electrostatic solvation energy of protein (J. Chem. Phys. 125, 09496, 2006)
Other application of the MFCC method:
Time-Dependent Quantum Wavepacket Dynamics for Chemical Reactions
Time-dependent wavepacket approach is a general and powerful computational
approach to study chemical dynamics problems. Recently, we developed xxx
Currently, Our group is focused on developing time-dependent quantum
wavepacket methods
as practical computational tools to accurately predict various dynamical
properties such reaction cross sections in gas-phase and gas-surface reactions.
Our ultimate goal is to develop computational methods that will allow us
to quantitatively predict reactivity for complex reactions in chemistry.
Related publications for
some triatomic reactions:
Related publications for
some tetraatomic reactions:
SVRT Model for Quantum Polyatomic Reaction Dynamics
We recently developed a SVRT (semirigid vibrating rotor target) model for
quantitative studying of reaction dynamics involving polyatomic molecules.
The SVRT model preserves the correct stereo-dynamics of the reaction system--
which is very important for polyatomic reactions. The main advantages of
the SVRT model are the following:
-
The model is quite general and applies, in principle, to a large class of
polyatomic or complex molecular reactions.
The
- The model automatically becomes exact for atom-diatom and diatom-diatom
systems.
- The mathematical dimension of the model is finite. For example, 4 for
atom-polyatom systems and 7 for general polyatom-polyatom systems.
- The model can be systematically improved further by including a selective
number of vibrational modes in the Hamiltonian.
Related publications on the
SVRT model:
Reactant-Product Decoupling (RPD) Approach
Related publications on RPD method:
Molecular Reaction on Solid Surfaces
Detailed quantum dynamics study for dissociative adsorption of hydrogen
molecule on metal surfaces shows folloing properties:
- Vibrational enhancement: Molecular dissociation is significantly
enhanced when the molecule is vibrationally excited.
- Steric effect: There is a pronounced steric effect in that molecular
dissociation strongly favors rotational states satisfying j=m or the
"Helicopter model".
- Selection rule: For dissociation of homonuclear diatomic molecular on
smooth or weakly corrugated surface or on high symmetry sites, molecular
rotational states with symmetry j+m=odd are significantly suppressed, and
states with j+m=even dominate dissociations at low energies.
Related publications on surface
reaction:
Ab Initio SOFA Quantum Dynamics
Related publications on SOFA method:
Direct Inversion of PES from Specstroscopic Data
Related publications on IPSVD method:
The Dynasol Project