4.2 The Resource Description Framework 63
As another example, consider a biologist who has just found a novel DNA
sequence from anAnophelesclone which may be important in the devel-
opmental process. To find related sequences, the biologist runs ablastn
search based on a set of requirements (e.g., the sequence identities must be
over 60% and the E-value must be less than 10−^10 ). These requirements can
be captured as rules and constraints which could be taken into account by
an online Semantic Web–enabled sequence comparison service. If the re-
searcher found a number of significantly similar sequences inDrosophila,the
scientist could then obtain gene expression data for the relevant genes from
a Semantic Web–enabled gene expression database. Rules can then be spec-
ified which capture the interesting expression profiles, such as genes which
are highly expressed at specified time points in the developmental process.
In both of these examples, the activities can, in principle, be carried out
manually by the researcher. The researcher reads material, selects the rele-
vant data, copies and pastes from the web browser, and then struggles with
diverse formats, protocols, and applications. Constraints and rules can be
enforced informally by manually selecting the desired data. All of this is
tedious and error-prone, and the amount of data that can be processed this
way is limited. The Semantic Web offers the prospect of addressing these
problems.
Summary
The Semantic Web addresses two important problems in Bioinformatics:
- The dramatic increase of bioinformatics data available in web-based sys-
tems and databases calls for novel processing methods.
- The high degree of complexity and heterogeneity of bioinformatics data
and analysis requires semantic-based integration methods.
4.2 The Resource Description Framework
One of the most basic mechanisms of computing systems and networks is
the ability to obtain information identified by an address of some kind. To
open a web page using a browser one must enter a URL either manually or
by clicking on a hypertext link. A URL identifies aresourcewhich can be a
whole web page, an item within a webpage or a query for information that
can be obtained at a website. The ability to access resources using a URL is a
fundamental service provided by the Internet.
