Biomolecular processes take place in complex environments. Our group
develops theoretical and computational methods to cope with this complexity
in order to understand some biological questions.
Of particular interest are enzymatic reactions in solution
and macromolecular interactions in cell.
Our research includes development of
combined quantum mechanical and molecular mechanical (QM/MM) methods for processes involving
changes in electronic structure. A significant effort is devoted to
the development of a full quantal force field, based on the explicit
polarization (X-Pol) theory.
Additionally, we develop valence bond-based techniques for condensed-phase
and biochemical transformations, making use of block-localized molecular
orbital theory (also known as Block-Localized Wavefunction or BLW)
and block-localized density functional theory (BLDFT). VB-like configurational
state functions can be constructued with BLDFT, which are used
in a multiconfigurational approach, called
multistate density functional theory (MSDFT). Unlike the traditional
Kohn-Sham DFT (KSDFT), the ground-state density is not used to determine the
We make heavy use of the computational resources at the University of
Minnesota Supercomputing Institute.
The following pages provide you with a brief description of some
of the research areas and projects
that have been investigated in our group.
If you would like to receive reprints of publications or
additional information concerning our research, you may direct your requests to
jiali at jialigao.org.
X-Pol: A New Paradigm for Biomolecular Simulations
|Molecular mechanics dates back to the pioneering studies of steric effects
independently by Hill and by Westheimer, whereas the force field for biomolecular
simulations was established by Lifson in the 1960s. Tremendous progress has been
made in the past fifty years; however, a sobering fact is that the basic
functional form, including polarization terms, has hardly changed. A fundamental
change in force field development is warranted in order to increase the
predictability of computational biology.
In the X-Pol potential, the system is treated explicitly by
electronic structure theory and the wave function (or electron
density) is optimized by self-consistent field (SCF) method.
The internal energy terms and electrostatic potentials
used in classical force fields are replaced and described
explicitly by electronic structure theory. Naturally,
electronic polarization and charge transfer are treated
by the theory. Furthermore, such a method can be
directly used to model chemical reactions, electron transfer, and
electronically excited states.
MSDFT: A Novel Multistate Density Functional Theory
|In collaboration with Professor Yirong Mo,
we recently introduced a multistate density functional theory (MSDFT)
in the framework of valence bond approach
for studying electron transfer and chemical reactions.
The method is based on a block-localized density functional theory
(BLDFT), first described by Professor Yirong Mo, for the
construction of valence bond-like electronic configurations,
or constrained DFT.
Since DFT is used to determine the valence
bond multistate Hamiltonian matrix elements, this method is called
multistate density functional theory (MSDFT) or VBDFT.
MSDFT is an extention of the mixed molecular orbital and valence
bond theory, called MOVB, which we introduced in 2000.
The key features of MSDFT include that the electron density of
the adiabatic ground state is not directly computed nor used to obtain the
ground-state energy, rather, it is
obtained by diagonaling the multistate valence bond Hamiltonian.
MSDFT represents a departure from the standard
single-determinant Kohn-Sham density functional theory.
Because MSDFT is a multistate multiconfiguration theory, it has the advantage of
including both dynamic (through DFT) and static (through
VB theory) correlation effects, providing a viable approach
for certain systems and reactions to
overcome self-interaction errors in current approximate KS-DFT.
Significantly, the BLDFT method is a versatile theory
that can be used to analyze conventional KS-DFT results
to gain insight into chemical bonding properties, including
intermolecular interaction energy decomposition analysis.
Biomolecular Interactions & Catalysis
|We are applying combined quantum
mechanical and molecular mechanical methods both at the semiempirical and ab initio level
to enzymatic processes in solution. At present, we focus on three fundamental areas of
biological interest: (1) mechanisms of enzymatic reactions including phosphate transfer
processes and carbocation cyclizations, (2) electronic and chemical transformations at the
electronic excited states, and (3) vibrational energy relaxation and dynamics of
substrate-protein interactions in the enzyme active site.
Dynamics & Interactions
Structure, Reactivity and Solvation
|Solvent effects have profound influence on
chemical reactions and reactivity. In many cases the direction of a reaction may be
altered by changing the solvent. The information gained in computer simulations may be
used in rational design of catalytic agents for organic synthesis. Several reactions are
currently being investigated, including photoisomerization, pericyclic and nucleophilic
addition/substitution reactions. We have recently developed a method, combining features
of molecular orbital and modern valence bond theory, to investigate the resonance and
stereoelectronic effects in organic molecules.
|Another exciting area of research is microporous materials and
catalysis. In particular, the adsorption, binding and reaction mechanism associated with
catalytic processes in zeolites are being studied. Microporous materials such as zeolites
are powerful industrial catalysts and have a wide range of applications. Computational
methods developed in our laboratory are capable of providing answers at the molecular
level to numerous questions of chemical and industrial interest. An understanding of the
catalytic mechanism will enable chemists to design and synthesize more powerful catalysts.
|Undoubtedly, the key factor that determines the success
of condensed phase simulations is the availability of accurate intermolecular potential
functions. Traditionally, empirical MM potentials are used; however, the form of these
functions are not appropriate for describing chemical reactions. The combined QM/MM
approach takes advantage of the accuracy and generality of QM calculations but still
retains the computational efficiency of the MM force field by treating the solute
quantum-mechanically and the surrounding solvent molecules classically. The use of QM
methods in statistical mechanical Monte Carlo and molecular dynamics simulations allows us
to simulate chemical processes in solution. Current interest includes development of
algorithms for even more accurate QM/MM calculations, a mixed molecular orbital-valence
bond (MOVB) approach for simulating chemical processes, and molecular orbital-based
polarization force fields..