Grid Computing

Abstract

Structural biology, in general, and NMR, in particular, have always been associated with advanced computing. The multistep nature of the pipeline from NMR data acquisition to resonance assignment, and finally to structure determination and analysis, has stimulated the development of a large number of programs that, in most cases, are designed to specifically address only one or a few of the steps. In addition, specific programs have been developed over the years to perform tasks based on bio-NMR data that are not in the structural determination pipeline, such as the analysis of protein backbone dynamics. This has resulted in a significant fragmentation of the computational aspects of bio-NMR, forcing scientists in the field to use a panoply of different, poorly interoperable software tools. The new challenges posed by the postgenomic era, which called for a much more extensive application of bio-NMR to a larger than ever variety of different molecular systems, stimulated initiatives to overcome these bottlenecks. A collaborative project entitled e-NMR was launched in this context; its aim was to exploit the powerful European Grid infrastructure, which enabled distributed computing power and storage sharing across geographical and administrative domains, to build a uniform environment that allowed users to apply a variety of computational tools to their bio-NMR data in a seamless manner, with continuously updated programs and exploiting on-line facilities for data shuffling among programs. The e-NMR initiative is now being continued by the WeNMR project. © 2012 Wiley-VCH Verlag GmbH & Co. KGaA.