user login 

BITS2007 Meeting
BITS2007 Meeting

26-28 April 2007 Napoli, Italy

email support
Home > Contribution details
get PDF of this contribution get XML of this contribution get ICal of this contribution
Gene Expression database as a tool to identify potentially interesting Oncological Targets
Data on gene differential expression between normal and tumor 
samples can be used to
get insight about putative physiological gene function, identify 
overexpressed in tumor vs normal cells or make hypothesis about 
possible side effects
of target inhibition.Unfortunately systematic information at protein 
level is usually
not available. Protein Atlas ( is a very 
relevant source of
information but so far it covers only ~1300 proteins.Gene 
expression levels, analysed
by Microarray (MA) technology, can partially compensate for the 
gap, giving
information at mRNA level but for the whole genome.
Thanks to the effort of public consortiums, thousands of MA 
experiments have been
collected in dedicated databases (ArrayExpress, GEO, ...), although 
they are not
organized in a way easy accessible for data mining and 
comparison.An attempt to
create a user-friendly interface for expression data in normal tissues 
as well as
cell lines has been done by Novartis (Novartis Gene Expression 
Atlas), with some
limitations. Gene Expression Atlas is a database collecting MA 
expression data
performed on cell and tissue samples coming from various 
commercial sources
(Clinomics Biosciences, AllCells, CloneticsBioWhittaker, ...) provided 
with an interface
to search for gene expression on a gene by gene basis. It includes 
several cell lines
and is easy to use but tissue source is not consistent and is poorly 
donor information is not available and only few tumor samples are 
included (mainly
leukemia). Moreover samples are pooled and performed in duplicate.

Searching GEO database we noticed two interesting data collections: 
GSE3950 (~300
chips from 65 diverse normal tissue types) and GSE2109 (~1500 
chips from ~1500
different patients with several tumor types). In both studies a 
particular attention
was dedicated to standardize the procedure for data collection and 
storage. On all
the samples microarrays were performed using Affimetrix technology 
The objective of the present work was to create a database to store 
gene expression
data together with available clinical information form the two dataset 
and provide an
interface to query the database and comparing different 
conditions.For this purpose a
database was created in Oracle and interfaced with Spotfire.

The database allows the comparison of gene expression level across 
displaying histograms of normal and tumor samples. Expression data 
are directly
linked to clinical information derived from patients (histology, grade, 
sex, relaps,
treatment type, .... Filtering and sorting options allow to focus on a 
specific tissue
of interest and to make hypotesis about gene expression profile and 
histology (ie.
adenoma vs. carcinoma). Heatmaps and gene profiles are also 
possible to display.The
database can be a starting point to identify potential new targets 
based on
differential expression profile tumor vs normal
Id: 111
Place: Napoli, Italy
Centro Congressi "Federico II"
Via Partenope 36
Starting date:
-- not yet scheduled --   
Duration: 01h00'
Contribution type: Poster
Primary Authors: BOSOTTI, roberta (Genomics Unit, Biotechnology Department, Nerviano Medical Sciences S.r.l.,Nerviano (MI))
Co-Authors: UKMAR, giorgio (Genomics Unit, Biotechnology Department, Nerviano Medical Sciences S.r.l.,Nerviano (MI))
CALOGERO, raffaele (Genomics and Bioinformatics Unit, Dipartimento di Scienze Cliniche e Biologiche c/o Az. Ospedaliera S. Luigi, Orbassano (TO))
ISACCHI, antonella (Genomics Unit, Biotechnology Department, Nerviano Medical Sciences S.r.l.,Nerviano (MI))
Presenters: BOSOTTI, roberta
Included in session: Poster Session
Included in track: Biomedical informatics | Last modified 08 July 2009 10:35 |

Powered by Indico 0.90.3