Grid Security - Grid Computing: Infrastructure, Service, and Applications - page 315

The life sciences in the post-genomic age offer huge potential benei ts
to mankind. At the same time, there is considerable concern regarding
the use of these datasets for other, potentially nonclinical uses. Insurance
agencies or employers would be keen to know the likelihood of those
they are offering policies/jobs to are of developing a chronic disease.
Public association with certain diseases or medical histories more gener-
ally leaves individuals rightly concerned that their data are not accessible
to the masses.
Within this context, the National e-Science Centre (NeSC) [25] at the
University of Glasgow has been exploring the development and manage-
ment of a wide range of e-Research infrastructures. At the heart of these
infrastructures are i ne-grained security, usability for the end user, and
support of interdisciplinary e-Research. Development of a new drug or
exploration of a given disease requires a wide array of researchers and
research domains to be involved. This can include biologists, bio-
informaticians, clinicians, statisticians, chemists, physicists, pharmacolo-
gists, and epidemiologists, among numerous others. Only through their
successful inputs can the whole research process be supported. To explore
how we have developed infrastructures supporting interdisciplinary
research, we describe several completed and ongoing projects, including
the MRC-funded Virtual Organisations for Trials and Epidemiological
Studies (VOTES) [26] project, the BBSRC-funded Grid Enabled Microarray
Expressions Proi le Search [27] project, and the DTI-funded Biomedical
Research Informatics Delivered by Grid Enabled Services (BRIDGES)
project [28].
For demonstration purposes we outline how this infrastructure can
support research into cancer. However, the same infrastructure can (and
is) be used across a wide range of other clinical areas. We also note that
the following are based upon representative datasets only.

12.5.1 VOTES Project

The VOTES project was funded by the MRC to develop a grid framework
through which a multitude of clinical trials and epidemiological studies
could be supported. Thus, rather than engineering bespoke solutions for
a given trial or study, VOTES focused on providing an infrastructure
where a multitude of trials and studies could be developed and supported,
each with their own particular nuances in terms of the data that are being
accessed, by whom, and the security policies that apply, and so on.
At the heart of clinical trials are three key processes: patient recruit-
ment, data collection, and study management. Recruitment is primarily
concerned with identifying patients who are potentially suitable for a clini-
cal trial, or even whether sufi cient patients exist who meet the criteria for
a given trial to take place, as might be the case when conducting a feasibi-
lity study. Once identii ed it is essential that patients are advised about all