Home » Technology » Scientific Computing » More On DNA Simulation

More On DNA Simulation

© Adam Hughes

This week, we continue our survey of current and recent work in biological simulation, with particular emphasis on calculations involving DNA molecules. This time around, out journey takes us the Physical Chemistry Department at Stockholm University in Sweden. This division has produced several recent studies which blend traditional chemistry simulation methods with the complexity of DNA.

Given the large numbers of atoms involved in a simulation of DNA and its surrounding medium (often water), it is little wonder that the researchers in Stockholm developed a parallel molecular dynamics, or MD, code to help lighten some of the real-world time burden. Indeed, the goal of this project was to develop a code that could simulate systems with 1000-10,000 atoms in a reasonable amount of time. Realizing that the force calculations are the most expensive (costly in terms of time required to complete) of any MD simulation, the scientists focused there efforts in this area. Their work was rewarded in the form of calculation speedup when more than one processor was used. In fact, in the simulation of a 635-atom DNA fragment surrounded by 1050 water molecules and 20 sodium ions, 32-processor simulations provided a speed-up factor of ten over the one-processor time. Considering the overhead involved with such a parallelization scheme, this is an impressive result.

Several other recent studies involving DNA have also been conducted by the scientists in Stockholm. Many of the simulations have involved the use of Monte Carlo techniques, another standard classical simulation tool. Some of the work here includes examining how different ions (charged atoms, in this case) compete with each other to bind to DNA molecules, the development of effective potential terms to be used in DNA simulation, and other study into the nature of DNA in various situations. Some of these studies are undoubtedly important stepping stones in the path toward more sophisticated DNA simulation, while others provide valuable insight about how DNA behaves in certain environments.

Be sure to visit the researchers' web page at

http://www.fos.su.se/physical/sasha/dna_...

By clicking on either author's name at the top of the page, you will be taken to his personal page, where you can learn more about his research interests and view some of his papers. The department homepage at

http://www.fos.su.se/physical/

is also chock full of information that you may be interested in.

The copyright of the article More On DNA Simulation in Scientific Computing is owned by . Permission to republish More On DNA Simulation in print or online must be granted by the author in writing.