Download the RbDe web service - Oxford Academic

Protein Engineering vol.13 no.6 pp.395–396, 2000
COMMUNICATION
Interactive construction of residue-based diagrams of proteins: the
RbDe web service
K.Konvicka, F.Campagne1 and H.Weinstein
Department of Physiology and Biophysics, Box 1218, Mount Sinai School
of Medicine, 1 Gustave L. Levy Place, New York, NY 10029-6574, USA
1To
whom correspondence should be addressed.
We announce the Residue-based Diagram Editor (RbDe)
web service that allows online construction of residue-based
diagrams and the creation of stored diagram libraries. The
service has been tuned for the construction of snake-like
diagrams (for transmembrane proteins) but can be used
to render any protein for which defined secondary structure
data or hypotheses are available. RbDe is freely available
through the Internet from our web site: http://transport.physbio.mssm.edu/rbde/RbDe.html. Licenses for intranet
uses can be obtained upon request.
Keywords: membrane proteins/residue-based diagrams of proteins/web service
Introduction
Residue-based diagrams are a special kind of representation
of a protein where residues are laid out on a page with
conventions that highlight the relations between a given
sequence, secondary structure information and residue-level
annotations. For transmembrane proteins, where the representation is used extensively as a simplified model because of
the lack of better structural information, the term snake-like
diagram has been coined, inspired by the layout of the sequence
that spans the membrane several times.
The automation of the rendering of snake-like diagrams, for
proteins whose topology was similar to that of G proteincoupled receptors (GPCRs), was first published with the
description of the Viseur program (Campagne et al., 1995–
97,1999). This first automation was important to database
curators in the GPCR field as it allowed them to build snakelike diagrams of GPCRs, designed to present a specific aspect
of a system. The GPCRDB database (Horn et al., 1998) uses
such snake-like diagrams, built with the Viseur program, in
order to (i) provide easy access to mutation data, each sequence
in the database being associated with a snake-like diagram
that shows mutated residues in a given color, and (ii) visualize
correlated mutation analysis results (Wells, 1990).
A new method (Campagne and Weinstein, 1999) was
developed to extend the automation of the rendering of residuebased diagrams to other kinds of topologies and to provide
more flexibility in the control of the output than was possible
with the first algorithm. We present here the Residue-based
Diagram Editor (RbDe) web service that provides the interface
for end-users to construct diagrams interactively for a sequence
of interest.
Organization of the service
The service consists of the following parts :
© Oxford University Press
Import facility. Allows the user to import a sequence into the
service. SwissProt sequences are transparently obtained
through the network and other sequences can be typed in
as text in a short list of sequence formats (SwissProt, PIR,
Fasta, PDB). The secondary structure and transmembrane
limit annotations are read when available in the given
format. Disulfide bridges are read from SwissProt files to
be represented on the diagram.
Topology editor. Provides a way to enter or refine the
secondary structure and transmembrane limits hypotheses.
Secondary structure elements can be chosen between helices
and strands.
Annotation editor. Manages annotations attached to a residue.
The annotations consist of the choice of a color and/or URL.
Individual residues and ranges can be linked to end-user
documents that describe an aspect of the system to which
they relate.
Diagram editor. Manages parameters that affect the rendering
of the diagram as a whole. Currently, these parameters are
direction of the N-terminal segment on the page, reduction
factor to be applied to the diagram and coloring of the
residues which are not annotated.
Diagram viewer. Presents the diagram as customized by the
user and enables the links defined in the annotation editor.
Diagram library. Allows end-users to store the diagrams they
created with the service. Stored diagrams can be retrieved
to be completed (addition of annotations) or simply consulted. (The library currently enforces that only the creator
of a diagram can consult and change it.)
Figure 1 presents the sequence of Connexin50 (SwissProt
ID: CXA8HUMAN) where the location of the residue 87
(proline) is identified. Mutation to a serine at this locus causes
a congenital cataract (Shiels et al., 1998).
Implementation
RbDe has been implemented as two Java packages that
make use of the Servlet Application Programming Interface
(JavaSoft, 1999a). The first package (edu.mssm.crover.scentral)
regroups features (e.g. user authentication, repository of URLs,
page templates, pool of database connections, basic support
for persistence, etc.) that are likely to be shared among
additional web services to be developed in the future. The
second package (edu.mssm.crover.rbdeservice) contains the
Servlets we use to construct the user interface. The implementation of edu.mssm.crover.domain2d, described recently
(Campagne and Weinstein, 1999), is used to build the diagrams
as directed by the parameters edited through rbdeservice. The
diagram library is implemented via the JDBC Application
Programming Interface (JavaSoft, 1999b) that provides access
to a relational database.
Conclusion
The RbDe web service can be used to build residue-based
diagrams of proteins online. The service provides significant
395
K.Konvicka et al.
Fig. 1. Connexin50’s snake-like diagram. The annotation and diagram editors are shown at the top. Proline 87 is shown in white and could be hyperlinked to
the Medline abstract to document the associated disease phenotype.
flexibility in the construction of the diagrams and their annotations. Experience with the use of mutation databases (e.g.
TinyGRAP) indicates that RbDe should be attractive resource
for experimentalists who need these kind of diagrams to
summarize and communicate results as well as plan future
experiments. The diagrams built with the service can be
reused programmatically to present the result of computational
analysis (for instance, to present the location of predictedturns on the sequence, etc.). Documentation of the Application
Programming Interface necessary for doing so is available online at http://transport.physbio.mssm.edu/joe.
Campagne,F., Jestin,R., Reversat,J., Bernassau,J. and Maigret,B. (1999) J.
Comput.-Aided Mol. Des., 13, 625–643.
Campagne,F. and Weinstein,H. (1999) J. Mol. Graph. Mod., 17, 207–213.
Horn,F., Weare,J., Beukers,M., Horsch,S., Bairoch,A., Chen,W., Edvardsen,O.,
Campagne,F. and Vriend,G. (1998) Nucleic Acids Res., 1, 275–279; see
also URL: http://www.gpcr.org/7tm.
JavaSoft. (1999a). Java Servlet API. URL: http://java.sun.com/products/servlet/,
Dec.
JavaSoft. (1999b). JDBC API. URL: http://java.sun.com/products/jdbc/, Dec.
Kristiansen,K., Dahl,G.S. and Edvardsen,O. (1996) Proteins: Struct. Funct.
Genet., 26, 81–94.
Shiels,A., Mackay,D., Ionides,A., Berry,V., Moore,A. and Bhattacharya,S.
(1998) Am. J. Hum. Genet., 62, 526–532.
Wells,J. (1990). Biochemistry, 29, 8509–8517.
Acknowledgement
Received January 28, 2000; revised March 15, 2000; accepted March 22, 2000
This work was supported by National Institute of Health grant DA12408.
References
Campagne,F., Bernassau,J.-M. and Maigret,B. (1995–97) The Viseur Program.
Laboratoire de Chimie Theoretique de Nancy, Nancy. WWW: http://
transport.physbio.mssm.edu/viseur/viseur.html. Viseur Project home page.
396