Transcript Diapositive 1 - LBGI Bioinformatique et Génomique Intégratives

ImAnno: a web-based tool to annotate
tissue-specific gene expression patterns
Laura Cammas1, Guillaume Berthommier2, Raymond Ripp2, Pascal Dollé1
1 Component B, Departement of Physiological Genetics
2 Component T, Laboratoire de BioInformatique et Génomique Intégratives
IGBMC (CNRS UMR 7104, INSERM U 596), CERBM, 1 rue Laurent Fries, Illkirch 67404, Strasbourg France
Abstract.
High throughput analysis of gene expression generates large amounts of data that need to be analyzed and integrated. GenePaint.org is a digital atlas of gene
expression patterns for thousands of genes in the E14.5 mouse embryo. The data, freely available on internet, are produced by large scale automated in situ hybridization
(ISH) and are currently being annotated in another database (EURExpress.org). Our aim is to take advantage of this unique resource to identify novel genes specifically
expressed in eye structures. We propose to systematically implement our analyses (‘annotations’) of gene expression patterns to the EVI-GENORET database. For this
purpose, the ImAnno Genoret Image Annotation Tool was developed. It allows the rapid annotation of all types of images through a user friendly web interface that can be
made available to all EVI-GENORET members. The annotations are stored in the Genoret Database and can be retrieved and queried by the Genoret search engines.
The goal of this work is to identify new candidate genes whose expression is specific of the eye and perform large scale bioinformatics analysis to define gene families and
pathways that may function during eye development.
General overview of ImAnno. An annotated database for ISH expression patterns in E14.5 developing mouse eye
1
The form is organized as a hierarchical list of anatomical terms. The list
of tissue types is defined by the user(s) and can be managed
dynamically in order to add new tissue types in the course of the project.
2
http://www-genoret.u-strasbg.fr/genoret/ImAnno
After interpretation of the in situ hybridization images, the user completes the
annotation form by defining the expression level in the selected tissues.
General information
Currently we have included :
- 14 eye tissues,
- 5 non ocular head tissues ,
- 2 other body regions.
such as the gene name, symbol
and description, the accession
number and useful links to the
GENORET database are
automatically stored.
Schematic of the expression
domain and intensity
The level of expression is
represented by a color code for
rapid and intuitive evaluation.
3
Link to a
reference image
These annotations are processed using bioinformatic methods.
General expression pattern
staining
annotation
 Gene searching
eye
other
tissues
all negative
 Clustering based on expression pattern
 Automated bibliographical analysis

eye negative
ubiquitous

eye specific


Signal intensity
3 levels of expression are defined:
negative, weak or strong
4
The annotations will be released on the GENORET Database in the form of
‘ID cards’ (screens) for each gene allowing easy reading of the queried
gene expression pattern.
Search tools will be developed to query ImAnno for genes expressed in
particular organs or structures of interest, synexpression groups, etc.
Results.
To date 902 genes have been
annotated using ImAnno. Of
these, 119 show specific
expression in one or several eye
tissues. Some of them, like Pax6,
are well known but interestingly
this aproach allowed to identify
genes not previously known to be
expressed in the eye.
Some structures can be absent
from the available images, and
are therefore “not annotated”
Characterization of the
expression (regionalized, etc)
The ImAnno tool can be adapted to all kind of projects
13%
Eye Specific
Eye Negative
All Negative
Ubiquitous
35%
32%
20%
• From images to textual data
• It offers a common organisation and is easily configurable
• It is easy to use
 through a web interface
 available through all web browsers
 by several people (history and ownership)
• Automatic storage & indexation in the Database
 relational database
 standardisation
• Quering tools
 web display
 accessible for programs
 from inside or outside
• GENORET user can annotate and store other sets of ISH or IHC data
 modified ImAnno forms have been derived for
other projects in-team, in order to store information
on expression patterns in inner ear or tooth buds.
 the Eye fundus images ImAnno tool is already implemented and ready to use.