Biomedical Engineering Reference
In-Depth Information
(iv) Develop New Custom Capabilities in Database for Users Generally
each group may need some custom features developed, which greatly
facilitates their use of the database. This benefi ts the community as
a whole and also moves certain development tasks up the prioritiza-
tion list.
Several examples of capabilities we think will be essential for such databases
as CDD in future are listed in the next sections.
21.6.1.1 Develop an Ontology to Facilitate Advanced Data Mining Across
CDD Databases The biomedical research community, and specifi cally those
involved in neglected disease research, is generating very large data sets facili-
tated through HTS [42, 44, 45], and this presents impending informatics chal-
lenges both for selection of hit compounds for follow-up studies as well as
computational analysis of such data. A highly effective concept for ensuring
such data have continued utility and accessibility is through a formal ontology;
for example, the Gene Ontology [46, 47] and the OBO Foundry have success-
fully demonstrated the utility of this approach. The benefi ts of ontologies have
been well articulated by others and used to enable network analysis [51],
facilitate translational bioinformatics [52], and link diseases to animal models
[53]. Bioassay data are particularly well suited for management within an
ontology because they encompass a wide diversity of experimental designs but
usually have a limited range of prescribed objectives. A functional ontological
framework will allow new assay descriptions to be meaningfully integrated
into the knowledge base with relative ease.
Adoption of an open-assay ontology will be a major milestone in converting
volumes of assay data into machine - interpretable knowledge and fi nally human
insight. Given CDD's unique position with an already engaged research com-
munity, the key prerequisites are in place to make the ontology widely adopted
and therefore maximally useful.
Incorporation of an ontology which captures this complexity will allow for
more precise information extraction and integration of the various structured
and unstructured data sources. These ontologies will be incorporated into the
fabric of CDD with strategic “push” and “pull” mechanisms to promote adop-
tion. Once a fi rst pass is made at automated assay annotation, a series of simple
questions (Boolean or short list choices) will complete the entry and ensure
accuracy of the automated procedures.
CDD will assist the community in annotation of its assay defi nitions and
data within an open-assay ontology. Because of the large backlog of historical
assay instances and the importance of demonstrating a benefi t quickly to
promote adoption, CDD will directly assist partners in assay annotation.
Through this assistance, project partners will greatly accelerate the completion
of the task and ensure accuracy and fi delity.
An ontology will facilitate collaboration between researchers in a public/
private data-hosting environment by enabling automated systems to alert
Search WWH ::




Custom Search