Biology Reference
In-Depth Information
server retrieves the relevant information from the database via the API;
this information is then used to dynamically generate a web page for the
user.
33
The user can view the genome in various ways, either scrolling
along an entire chromosome and zooming in and out or as a map of
genetic markers, genes, and proteins. The aim of the Ensembl genome
browser is to provide easy access to specifi c information, to provide a
high-level overview, or to provide a way to examine patterns in the
data. As such, a great deal of attention is paid to the details of the dis-
play: the organization of information on the page, the speed at which it
loads, and the colors and shapes used to represent various objects were
constantly under discussion among the Ensembl team. During the time
I visited, the team released an updated version of its website. The design
changes they described focus almost exclusively on visual issues:
One of the continuing challenges with a site that contains so
much data is making it discoverable. In testing sessions we
found that users were overlooking the left-hand menu with its
links to additional data, perhaps because there was so much on
the page already that they didn't expect to fi nd even more! . . .
The main change we have made is to improve the way that us-
ers move between pages. The displays have been grouped into
Location, Gene, Transcript and Variation and tabs at the top of
the page allow the user to switch easily between these sets of
views. If you browse to a location on a genome, initially only the
Location tab is present, but the others appear as you select spe-
cifi c genes and transcripts to view. . . . You will also have noticed
that we have changed the colour scheme somewhat, darkening
the blue background and extending it to the whole page header,
and using the yellow background to draw attention to impor-
tant parts of the page such as genomic images and information
boxes.
34
The redesign was not just an effort to make the site visually appealing:
the Ensembl bioinformaticians realized that how they displayed the ob-
jects would exert a determinative infl uence on whether and how partic-
ular data elements were used and understood by laboratory biologists.
Deciding how to represent biological objects had important conse-
quences for what those objects were and how they could be used.
At roughly the same time as Birney and his team, biologists in the
United States also began to develop tools for seeing whole genomes. One
of the most successful—the UCSC Genome Browser—was developed at
