Information Technology Reference
In-Depth Information
10.7 Related Research
This section compares our research with others' targeting similar objectives
in distributed computing (such as grid or cloud) environments. This section
clearly shows our major contributions to this field.
There are several research projects investigating how biomolecular appli-
cations, particularly molecular docking simulations, can be run on distrib-
uted computing platforms. Some examples of DCI-based molecular docking
simulations are detailed in this section. Most of these experiments are on
grid computing resources with rare exceptions currently for the utilization
of clouds.
Tantar et al. [ 14 ] gave an overview of current efforts on how large-scale
parallel computing is applied to molecular simulations. The authors are also
involved in the Docking@Grid project [ 15 ] that aims to define the optimal
deployment architecture for grid-based molecular docking simulations and
provide the accurate definition of the molecular energy surface.
The WISDOM project [ 16 ] is an international initiative to enable a virtual
screening pipeline on a grid infrastructure. WISDOM was the first large-scale
in-silico docking experiment on public grid infrastructure. The project has
developed its own meta-middleware that utilizes the EGI (European Grid
Infrastructure) production infrastructure and is capable of submitting and
executing a large number of jobs on EGI resources. Although the WISDOM
production environment is capable of submitting any kind of application,
the flagship application of the project is AutoDock.
Tantoso et al. [ 17 ] described a similar approach for Globus-based grids.
A  web interface has been developed by the authors to execute experiments
using AutoDock3.05 on target grid resources. A small workflow automates the
process, which includes the preparation of the receptor, creation of parameter
files, calculation of grid energy, and finally the actual docking of the molecules.
Cloud-based molecular docking environments are currently hard to find
in the literature. The only example we know about was written by Kiss et al.
[ 18 ]; the authors described the implementation of similar molecular docking
experiments on Windows Azure-based clouds. However, that implementa-
tion was closely coupled with the Azure infrastructure, and the user inter-
face is less flexible, making further improvements difficult.
There are also examples for the utilization of higher-level user interfaces
for molecular simulations, all based on grid computing infrastructure. The
Australian BioGrid portal [ 19 ] uses the DOCK [ 20 ] molecular docking software
for the simulations. This work is part of the Virtual Laboratory Project that aims
to utilize grid technologies for solving large-scale compute and data-intensive
applications in the area of molecular biology. The project uses the Nimrod
Toolkit and the World Wide Grid test bed [ 21 ] to conduct the experiments.
Search WWH ::




Custom Search