MODELING THE FORCED EXTENSION OF NICKED DNA

点击翻译按钮获取中文摘要

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. We request Teragrid resources for our theoretical study of the forced extension of nicked DNA chains. Nicked DNA chains, self-assembled from short oligonucleotides, have been recently studied experimentally by one of the co-PIs (Kersey et al., J. Am. Chem. Soc., 126:3038-3039, 2004). The behavior of the nicked chains during the forced extension was found to be remarkably different from that of continuous DNA (cf. Strick et al., Ann. Rev. Bioph. Biomol. Struct., 29:523-543, 2000; Harris et al., Bioph. J., 88:1684-1691, 2004). We seek to uncover the structural mechanism of the nicked DNA forced extension and, perhaps, learn the ways to program different extension regimes by manipulating the DNA sequence. The obtained knowledge may foster nanotechnology applications of nicked DNA chains and advance our understanding of key biological processes, such as DNA repair and replication. The extension of nicked DNA is modeled in a series of steered molecular dynamics (SMD) simulations (Isralewitz et. al, Curr. Opin. Struct. Biol. 11:224-230, 2001), using the classical MD software NAMD and AMBER. The model system consists of around 45 thousand atoms. The preliminary studies reveal several possible scenarios of structural changes of nicked DNA during the forced extension. Additional computational resources are needed to collect a sufficient statistical sampling of these scenarios, probe different extension velocities, and study different sequences of nicked DNA. We request 30,000 SUs on Teragrid machines. Using NAMD benchmarks (http://www.ks.uiuc.edu/Research/namd/performance.html), we estimate that our typical SMD simulation of 10-20 nanoseconds (ns) requires about 4,000-8,000 SUs when run on 16 to 64 CPUs of Cray XT3; only 1,000-2,000 SUs would be required on a faster BlueGene/L machine. Therefore, our request amounts to 5-8 additional SMD runs on slower Teragrid machines, allowing us to approximately double our presently accumulated data sampling at the probed extension velocities. On faster Teragrid machines, the allocation would be equivalent to 20-30 SMD runs, which would enable us to also probe another DNA sequence and repeat the original simulations with a different forcefield. In case of encouraging results, a more extended proposal for continuing the studies will be submitted to MRAC.