Biology Reference
In-Depth Information
with the dictyBase staff, we decided to consolidate all
D. discoideum
UniProtKB entries at the level of TaxID 44689.
As a result, we have a complete proteome set in UniProtKB, and the
keyword “complete proteome” was added to the relevant entries. This
allows users to download a complete, nonredundant set of proteins from
a given organism by specifying the TaxID of the species of interest with
the concomitant presence of the “complete proteome” keyword. Hence,
13 018
D. discoideum
UniProtKB entries (out of 14 179) are now tagged
with the “complete proteome” keyword (release 14.0 of July 22, 2008).
They can be retrieved at the following URL: http://www.uniprot.org/
uniprot/?query=organism%3Adictyostelium+AND+keyword%3A%22
complete+proteome.
2.2. UniProtKB/Swiss-Prot Annotation
The UniProtKB annotation process encompasses the annotation not only
of the sequence itself, but also of sequence features, gene and protein
nomenclature, and protein functional properties. Basic steps are done in
all of these areas for the whole UniProtKB, but an entry will be inte-
grated into Swiss-Prot only after it has been completely annotated and
manually reviewed in all of these annotation aspects. TrEMBL records
only partially fulfill the criteria outlined below, and are not manually
checked.
Some 6500 publications specifically related to
D. discoideum
are cur-
rently referenced in PubMed and constitute the basic material for manual
annotation, both at dictyBase and at UniProt. Systematic BLAST analysis
7
allows one to find the closest annotated protein and thus classify the pro-
tein as a member of a given family. This step is very important to ensure
consistency of annotation between organisms within a family. Since the
D. discoideum
proteome is highly enriched in asparagine stretches, the
BLAST tool had to be configured to filter low-complexity regions. In cases
where BLAST does not allow one to unambiguously identify the family
or subfamily to which a protein belongs, further analysis is performed
using phylogenetic tools such as the PhyloFacts database, the team
of which helped us to select candidates for priority annotation during the
annotation jamboree
.
8
