Theoretical and computational biophysics is a fast-evolving research field of biophysics, building theoretical models and using multiscale computer simulations to study the physical properties (structures, kinetics and thermodynamics) and biological functions of biomolecules and their complexes. Recently, we are particularly interested in macromolecules DNA and proteins.
Representative works
Stress-induced sequence-specific transition of DNA
dsDNA in vivo is topologically constrained. In mesophilic organisms dsDNA is usually negatively supercoiled (to the extent of 5-8% of supercoil density), which imposes significant twisting force on dsDNA and hence destabilizes the paring and stacking between bases. This inevitably induces structural transitions on dsDNA. Note that DNA is a heterogeneous macromolecule (DNA sequence is composed of 4 letters, A, G, T, C) , the transitions are always sequence-specific. Among the various transitions, local melting of the two strands of DNA ( i.e., bubble formation) is the most common case and has an impact on the functioning of DNA, for instance, to facilitating DNA duplication and gene transcription. Based on a well-known model (Benham model) to describe the bubble formation, we developed a novel and fast algorithm to calculate the melting profile of any DNA sequence, and observed an ubiquitous phenomenon in stress-induced melting, the inhibitory competition among distinct melted local regions, which is referred as Long-Range Allosteric Transition (LRAT) of DNA. After a detailed analysis of some eukaryotic DNA sequences, we found that LRAT may play a very important role in gene transcriptional regulation.
Related publications:
[1] Ming Li* and Zhong-can Ou-Yang, Predicting the function of eukaryotic scaffold/matrix attachment regions via DNA mechanics, J. Phys.: Condens. Matter, 2005,17:S2853-S2860
[2] Ming Li* and Zhong-can Ou-Yang, An exact numerical method to calculate the base-unpairing probability for any given DNA sequence by Benham model, Thin Solid Film, 2006, 499:207-212
[3] Ming Li*, Zhong-Can Ou-Yang,DNA as active polymer: long-range allosteric effect and chromatin loop structure,Computer Physics Communications,2007,177:176–179
Walking mechanism of the motor protein kinesin-1
Conventional kinesin 1, the world’s tiniest biped porter, is one of the most intriguing and the best studied biomolecular machines. It is a protein homodimer, and each monomer consists of an N-terminal globular head domain, a stalk region responsible for dimerization, a neck linker that connects the head and stalk, and a C-terminal fan-shaped tail domain which can be linked to external loads. The N-terminal head domain contains a microtubule binding site, and a nucleotide binding site which can also catalyze ATP hydrolysis. Powered by the chemical energy of ATP hydrolysis, kinesin can perform a unidirectional and hand-over-hand walk along the microtubule for a long distance. How the chemical reaction couples to the walking and how the structure determines the walking behavior (e.g., direction, velocity, processivity ) are the central issues in the study of kinesin-1. The neck linker is believed to be one of the most important elements to shape the walking behavior. We have carried out several theoretical studies is that aspect. In particular, we studied how the neck linker length can significantly affect the stepping velocity of the motor, by using a highly simplified chemically powered ratchet model. We found that the length of the wild-type neck linker (_15 a.a.) might be optimally designed for kinesin-1 to approach the largest stepping velocity.
Related publications:
[1] Y.Shu, X. Zhang, Z.C. Ou-yang, and M. Li*, The neck linker of kinesin 1 seems optimally designed to approach the largest stepping velocity: a simulation study of an ideal model, J. Phys.: Condens. Matter 24 (2012) 035105
[2] Ming Li, Yaogen Shu, Zhong-can Ou-Yang,Mechanochemical coupling of kinesin studied with a neck-linker swing model, Communications in Theoretical Physics 51:1143-1148 (2009)
[3] Yaogen Shu, Ming Li*, Substeps of kinesin studied with a neck linker swing model, Modern Physics letters.B. 24(6) : 539–548 (2010)
Multiscale modeling and simulations
The understanding of complex life phenomena bases on the widely knowledge of individual behaviors of biopolymers and their interplays in cell, which have been very actively studied in experiments. Computer simulations of these macromolecules are expected to largely improve our knowledge in explanation of experiments and formation of theories. Due to very slow dynamics and the complexity of biologic macromolecules, the simulating researches are very initial even in single molecule or dilute solution of polymers. The most interesting properties of biological macromolecules are dependent on their local chemical details (such as, changes of key bases in DNA or RNA molecules strongly vary their global conformations and function). However, usual simulations with complete chemical details can not catch the long-time dynamics. For example, the present most simulations of protein folding are based on simple models without chemical details or only arrive at very short time. The situation is being improved by developing and applying multiscale modeling and simulating techniques in some particular systems. General multiscale methodology is also attempting. We focus on the development and applications of multiscale simulations to stude slow dynamics of DNA and proteins, including coarse-graining, enhanced sampling, accelerating molecular simulation, metastable state analyzing, sampling technique in path space etc.
Related publications:
[1] Shun Xu, Xin Zhou*, and Zhong-can Ou-Yang, Parallel Tempering Simulation on Generalized Canonical Ensemble,Parallel Tempering Simulation on Generalized Canonical Ensemble,Comm. in Comput. Phys. (in press),2012. 
[2] Linchen Gong, Xin Zhou*, Kinetic Transition Network based on Trajectory Mapping,Kinetic Transition Network based on Trajectory Mapping,J. Phys. Chem. B114, 10266,2010.
[3] Linchen Gong, Xin Zhou*, Structuring and Sampling in Complex Conformational Space: Weighted Ensemble Dynamics,Structuring and Sampling in Complex Conformational Space: Weighted Ensemble Dynamics,Phys. Rev. E80, 026707,2009.
[4] Xin Zhou*, Yi Jiang, Steen Rasmussen, and Hans Ziock, Bridging coarse-grained models by jump-in-sample simulations,Bridging coarse-grained models by jump-in-sample simulations,J. Chem. Phys. 128, 174017,2008
[5] Xin Zhou*, Yi Jiang, Kurt Kremer, Hans Ziock, and Steen Rasmussen, Hyperdynamics for entropic systems: Time-space compression and pair correlation function approximation, Phys. Rev. E Rapid Comm. 74, 035701, 2006.
[6] Xin Zhou, Denis Andrienko, Luigi Delle Site, and Kurt Kremer, Flow boundary conditions for chain-end adsorbing polymer blends, J Chem. Phys. 123, (2005) 104904-1 - 104904-6.
[7] Xin Zhou, Denis Andrienko, Luigi Delle Site, and Kurt Kremer,Dynamic surface decoupling in a sheared polymer melt polycarbonate melt sheared over a nickel surface, Europhys. Lett. 70, (2004) 264 - 270.