Biology Reference
In-Depth Information
A recent shift in data collection and representation is seen in the
move from bioinformatic databases to atlases (e.g. EMAGE,
57
Protein
Atlas,
58
Novartis Atlas,
59
PeptideAtlas
60
). For example, PeptideAtlas
and Protein Atlas have recently been introduced as resources for brows-
ing and querying protein data. In this manner, data exploration may
become an indispensable preliminary step towards integration.
Interestingly, the navigation between different views in an atlas might
help us refine the definitions required for the overall task of biological
data integration.
So far, Gene Ontology (GO)
61
remains the most popular initiative
for organizing biologically interpretable protein data at molecular
and cellular levels. But in GO, knowledge is currently unevenly repre-
sented due to the relative novelty of this effort and also to our limited
understanding of biology. Ontology design has recently become an
active field in bioinformatics,
62
and this trend further justifies the need
for the determination of principles that set a more stable basis for data
integration.
The strength of integration has been defined to distinguish a portal
(loose integration) from solutions based on global data models (tight
integration).
63
In the following, integrated resources are presented in
order of growing strength of integration.
4.2.1. Proteomics servers
ExPASy was created in 1993,
64
and is still developed and maintained at the
SIB.
15
ExPASy is dedicated to the federation of databases and tools that
are relevant to proteomics studies. The server hosts the following data-
bases, which are essentially developed and maintained by SIB members:
•
the UniProt Knowledgebase (UniProtKB; see Chapter 6);
•
SWISS-2DPAGE (described in Sec. 4.2.3);
•
the PROSITE database of protein domains and families;
•
the ENZYME repository of information relative to the nomencla-
ture of enzymes; and
