Principal Investigator Michael A. Collins Project s51
Research School of Chemistry Machine VP
Co-Investigators: Fei Zhang
Research School of Chemistry
Model Simulations of B-DNA Dynamics
While the primary structures (base pair sequence) of many segments of natural deoxyribonucleic acids (DNA) have been determined, little is understood about the dynamics of these double-stranded biopolymers. From molecular biology, we know that both replication and transcription of DNA involve at least local opening of the duplex; even at physiological temperature DNA may show large amplitude stretching of the hydrogen bonds which connect the complementary base pairs, and this transient opening is relevant to DNA-drug binding. Moreover, thermal denaturation of DNA starts by formation of open "bubbles", preceding unwinding of the whole helix. We study this "local opening of the duplex" using a simple model.
What are the basic questions addressed?
Can a relatively simple model of DNA reproduce known gross characteristics of the dynamics? In more detail, how does the the local base pair dynamics depend on the base pair sequence.
What are the results to date and future of the work?
We have used this model to simulate several DNA sequences for about 10 nanoseconds at temperatures ranging from 200K to 520 K. It has been shown that the model is able to reproduce some features of DNA dynamics. In particular,the helix structure of the DNA model is stable at low temperatures, but melts above a certain temperature, where the two strands of the DNA begin to unwind and separate from each other, with increasing numbers of disrupted hydrogen bonds. The melting transitions are found to be rather slow in time; it takes longer than ten nanoseconds to completely denature a DNA sequence of only 100 base pairs. At temperatures below the melting point, localized fluctuations of the hydrogen bonds are commonly observed, especially in regions enriched with AT pairs. In the future there will be further development of the physical model.
What computational techniques are used and why is a supercomputer required?
Classical trajectories are used to study the dynamics. To obtain reliable results for trajectory averages, it is necessary to model large segments of DNA (hundreds of base pairs) over relatively long periods (at least tens of nanoseconds); this requires a supercomputing environment.
Model simulations of base pair motions in B-DNA,
M. A. Collins and F. Zhang, Nonlinear Excitations in Biomolecules,
(Springer -Verlag, Berlin, ) 117-124(1995)
Dynamical disorder in a model of base pair motion
in DNA, M. A. Collins and F. Zhang,
Fluctuation Phenomena: Disorder and Nonlinearity, World Scientific,
Model simulations of DNA dynamics, F. Zhang and M.A. Collins,
Physical Review E, 52, 4217 (1995).