MOLECULAR DYNAMICS SIMULATIONS OF SITE-SPECIFIC PROTEI-DNA INTERACTIONS

这个子项目是许多研究子项目中的一个 由NIH/NCRR资助的中心赠款提供的资源。子项目和 研究者（PI）可能从另一个NIH来源获得主要资金， 因此可以在其他CRISP条目中表示。所列机构为 中心，不一定是研究者的机构。 我们被授予服务单位在TCS的形式，中等资源分配补助金，以解决假说的作用，限制性内切酶的特异性动力学。这项资助的目的之一是使用计算机模拟来研究混杂突变EcoRI内切酶与其DNA识别位点的相互作用。我们使用分子动力学（MD）模拟来询问模拟是否可以重现通过X射线晶体学观察到的突变体和野生型复合物之间相对细微的差异。事实上，MD轨迹证实了在实验结构中观察到的许多特征。计算数据还指出了两种复合物之间的一些新的差异，这些差异需要通过进行显著增加持续时间的模拟来进一步研究。当我们申请服务单元时，我们的请求量基于与我们大小相似的系统的已发布基准，在TCS上运行pmemd代码（Amber 8软件套件）。这些基准测试不包括文件i/o所需的时间，在写入TCS上的磁盘卷时，文件i/o是有效扩展的一个重大障碍。因此，我们的MRAC分配将用完之前，我们能够解决我们的模拟提示的一些有趣的问题。我们与PSC科学家Troy Wymore和Shawn Brown进行了深入的讨论，他们强烈建议研究在Cray XT 3上运行我们的系统模拟将提高扩展性和吞吐量的可能性。因此，我们希望在XT 3上分配服务单元，以延长我们当前模拟的持续时间，并探索这种新PSC资源的能力，以更有效地利用未来的超级计算机时间。

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 were awarded service units on the TCS in the form of a Medium Resource Allocation grant to address hypotheses regarding the role of dynamics in restriction endonuclease specificity. One of the aims of this grant was to use computational simulations to study the interactions of a promiscuous mutant EcoRI endonuclease with its DNA recognition site. We used molecular dynamics (MD) simulations to ask if simulations could reproduce the relatively subtle differences between mutant and wild-type complexes that were observed by x-ray crystallography. Indeed, the MD trajectories confirm many of the features observed in the experimental structures. The computational data also point to some novel differences between the two complexes that warrant further investigation by conducting simulations of significantly increased duration. When we applied for service units, we based the amount of our request on published benchmarks for a system with size similar to ours, running the pmemd code (Amber8 software suite) on the TCS. These benchmarks do not include the time required for file i/o which represents a significant barrier to efficient scaling when writing to disc volumes on the TCS. Thus our MRAC allocation will run out before we are able to address some of the interesting questions prompted by our simulations. We engaged in extended discussions with PSC scientists Troy Wymore and Shawn Brown who strongly recommended investigating the possibility that running simulations of our systems on the Cray XT3 will improve both scaling and throughput. We would therefore like to request a development allocation of service units on the XT3 as a means to extend the duration of our current simulations as well as to explore the capabilities of this new PSC resource to make more efficient use of future supercomputer time.