transferring site specific annotations will not be possible. The
user is advised to continue from Subheading2.3 following a
quick investigation in UniProt protein page of the similar
resulting protein, as described in Subheading2.2.
(c) If the identity value is below 90%, it is still possible to transfer
the generic functions but not the site-specific ones; as a result,
following the procedure from Subheading2.3 is advised.2.2 Recorded Protein
Annotation (UniProt)
In a UniProt protein page (e.g., human Mast/stem cell growth
factor receptor KIT protein -UniProt id: P10721- athttp://www.
uniprot.org/uniprot/P10721), sequence properties and annota-
tions are grouped under distinct titles (function, taxonomy, pathol-
ogy, PTMs, expression, interaction, structure, family and domain,
sequence, etc.). The left pane shows these titles for easy navigation
on the main part of the page. The sources of functional information
in UniProt are explained inNote 2.
Under the heading “Function,” various types of functional
information are listed. First of all, a paragraph made up of curated
free-text is given. This is a summary of relevant literature publica-
tions, references of which are given at the end of the paragraph.
Second, region and site specific information (e.g., catalytic, metal,
active, and binding sites) is listed in tables, together with the
respective positions of the corresponding regions/sites on the
sequence. Third, the annotated functions in terms of the Gene
Ontology (GO) system is given for the three main GO categories:
molecular function, biological process and the cellular component.
Clicking a functional GO term in this list opens up the details of the
corresponding functional term in the QuickGO web-service.
QuickGO is both a database and an easy to use browser for protein
GO annotations [21]. Furthermore, clicking “Complete GO anno-
tation...” button in the UniProt protein page will take the user to
the dedicated protein GO annotation page in QuickGO (e.g.,
QuickGO page of the human KIT protein:http://www.ebi.ac.
uk/QuickGO/GProtein?ac¼P10721).
Under “Family & domains” section, subheading “Phyloge-
nomic databases” displays the clickable links to the corresponding
pages in different phylogeny resources (eggNOG, GeneTree,
OMA, OrthoDB, etc.). Another subheading “Family and domain
databases” shows the information and links for the domains that
the corresponding proteins are known to—or predicted to—con-
tain, in various resources (InterPro, Pfam, SUPFAM, Gene3D,
etc.). Each domain/family entry in these databases is displayed
with a functional description and site-specific information
(if possible).
Near the top of the left pane there is the “feature viewer” button
and clicking on it will display the ProtVista tool to visualize all of the
annotated features of the protein (e.g., human KIT protein atPhylogenetics-Based Prediction of Functional Sites 55