WISDOM: A Grid-Enabled Drug Discovery Initiative against Malaria - Grid Computing: Infrastructure, Service, and Applications - page 351

will assist the scientist and the decision-maker in organizing their
work in a l exible manner, and in delivering the information and
knowledge to the organization.
Deploying intensive computing is a challenge for

• in silico drug

discovery. For instance, computing 1 million docking probabili-
ties or modeling 1000 compounds on one target protein requires
in the order of a few TFlops per day. Very large computing
resources are also needed to describe accurately protein structure
models by computational methods based on all-atom physics-
based force i elds including implicit solution. Computing power
is also required for bioinformatics resource centers where server
access is saturated by the large number of short tasks requested
by users.
Joining new information technologies with life sciences to enable •

in silico drug discovery requires strong remote collaboration
between different public and private experts when addressing
neglected and emerging infectious diseases. It also involves strong
sharing of resources: data and knowledge, software and workl ow,
and infrastructures such as computing, storage, and networks.
The collaboration space needs experts to maintain the resources.
Having tools and data accessible to everyone in collaboration
requires intuitive interfaces that need to be maintained. These
interfaces reduce the development time of new methods. They also
help the integration of data and software from in silico drug
discovery but also from experimental processes.
Security is a key challenge for pharmaceutical industries but also •

for academic institutes in most cases. Effective protection of intel-
lectual properties and sensitive information requires, for instance,
authentication of users from different institutions, mechanisms
for management of user accounts, and privileges and support for
resource owners to implement and enforce access control policies.

In summary, the main requirements to develop in silico drug discovery
are data and software integration, intensive computing deployment,
remote collaboration and resources sharing, and, of course, security. Thus,
there is a need for powerful and secured environment sharing and inte-
grating remote resources such as tools, data, computing, and storage.

14.2.1.2 Grid Added Value

The grid added value in the development of in silico drug discovery for
neglected and emerging infectious diseases has multiple dimensions:

• Grids offer unprecedented opportunities for resource sharing

and collaboration.