Molecular Dynamics Applications

One of the great challenges in Proteomics is the folding problem, which is the task to predict the protein native structure solely from the amino acid sequence. This has become a major research field as a consequence of the increasing amount of sequence data acquired by DNA analysis and various genome projects. The acquisition of sequence information is a fast and straightforward procedure.

On the other hand the experimental determination of the three dimensional structure assumed by a protein is a rather slow process. Therefore, algorithms capable to translate the linear information given by the protein sequence into the spatial information defining the structure are required. This software is very time consuming: a distributed approach, like the GRID computing, is suitable to contain the elaboration time.

Besides structure prediction and protein design, one of the goals of biocomputing is the study of function properties of biological molecules at the atomic level. Classical molecular dynamics simulations can be applied to a number of investigations, including the study of folding, misfolding and aggregation of proteins as well as the destabilization or structural change of a protein upon mutation of one or more amino acids. The distributed computing approach gives the opportunity of overcoming the traditional computational limit in such simulations. This approach allows the calculation of long term dynamic trajectories, leading to an exhaustive sampling of structures. On the other hand the systematic study of point mutations is also made possible by replications on large scale of this kind of simulations.

Molecular Dynamics will be performed notably using as a starting point the data generated by the WISDOM application on the EGEE infrastructure. The challenge of the Wide In Silico Docking On Malaria initiative is to demonstrate the relevance and the impact of the grid approach to address Drug Discovery for neglected diseases. WISDOM focuses on massive in silico docking which is a computational intensive task as a first step towards in silico drug design.