Biomedical Engineering Reference
In-Depth Information
component describes the protein target, pathway, biological process or event,
and so on, targeted by the assay. Format includes biochemical, cell based,
organism based, and variations thereof. Technology describes the assay meth-
odology, assay design, and implementation (including detection method) of
how the perturbation of the biological system is translated into a detectable
signal. Analysis describes how the raw signals are transformed into reported
endpoints. Endpoints are the fi nal HTS results as they are usually published
(such as IC
50
, percent inhibition, etc.). BAO also captures other assay proper-
ties and relationships, such as assay purpose or how assays are related in
campaigns. BAO is designed to handle multiplexed assays. All main BAO
components include multiple levels of subclasses and specifi cation classes,
which are linked via object property relationships forming a knowledge
representation.
All these aspects of the BAO project described above will facilitate collabo-
rations among scientists of various disciplines, including screening biologists,
chemical biologists, medicinal chemists, cheminformaticians, and modelers.
The bioassay ontology will enable scientists to readily compare screening
results and to evaluate screening outcomes in the context of the existing large
public data sets—for example to distinguish artifactual hits from desired ones.
It will make it much easier to retrieve quantitative outcomes and activity
profi les relevant for medicinal chemists. BAO will also enable modelers to
generate consistent quantitative data sets that are related to a distinct biologi-
cal process or mechanism of action, for example, a protein target, a pathway,
an assay technology, and so on. Because ontologies defi ne semantics using text
expression, BAO can also make domain-specifi c information accessible to
nonexperts, for example, chemical structural information to cell biologists or
assay technologies to cheminformaticians. With an ontology of suffi cient detail
and appropriate software tools, one could imagine being able to express
complex queries covering several concepts by simple text, for example, “pro-
miscuous inhibitors of luciferase reporter gene assays.”
This BAO will also facilitate collaboration between researchers in a public/
private data-hosting environment by enabling automated systems to alert
researchers of potential collaboration opportunities (see Chapters 12 and 21).
For example, the owner of a private assay instance might opt to have an intro-
duction made to researchers that meet certain criteria based on assay data (as
opposed to structure or other alerts). This can be done in an automated
fashion, without either party viewing confi dential information directly. An
assay ontology makes this possible because it allows considerable fl exibility in
defi ning the terms of the alert and can make it very easy to defi ne such an
alert (i.e., by textual expressions). The private assay instance owner might then
opt to suggest collaboration only with other researchers who meet certain
affi liation requirements and have agreed up-front to a standard confi dentiality
agreement (e.g., available from Science Commons or elsewhere). Once both
parties acknowledge the collaboration, the data set (and means of direct com-
munication) could be shared within whatever database software is used. This
Search WWH ::
Custom Search