ELOGE is a comprehensive software suite to support our microarray production activities. It is built around a robust and carefully thought out relational database system. The 20 gigabytes of data stored locally and represented in over 100 tables conceptually fall into three categories: genomic, clinical and expression profiling data.
The major part of the genomic data is
extracted from public databases (UNIGENE, GENBANK, SWISSPROT, IMAGE) supplemented with private data (clone libraries, ESTs etc).
Stringent data consistency checks as well as extensive database
cross-linking ensure a coherent genomic resource combining primary data
(sequences, libraries & clones) with curated biological annotation (gene
symbols, protein functions, ontologies etc).
Clinical data are represented
through specific objects (such as patients, disease classification,
biological tests, therapeutic protocols etc) developed with our medical
partners.
The experimental data acquired through expression profiling
experiments are stored both as raw output as well as normalized and
consolidated overall measures. Each quantitative expression measurement
point is internally linked to its relevant qualifiers: underlying gene
identity and
sample characteristics. The system has been developed from the start to be
fully MIAME compliant (MGED) and will support MAGE-ML data exchange as soon as
final XML specifications are adopted.
This data warehouse constitutes the knowledge base upon which independent
and complementary software modules operate. The modules cover the two main
production phases in cDNA based microarray experiments.
Pre-Production: the
GENEFINDER and CLONECHOOSER software modules support the design of specific
microarrays. The GENEFINDER is a biological annotation data mining tool
aimed at establishing the sets of relevant genes to monitor. In order to
optimize the representation of these genes on microarrays, the dedicated
CLONECHOOSER software module fully automates and standardises the cDNA
clone selection process.
Production: the MicroArray Facility module (MAF) is a complete LIMS currently being developed in
support of our microarray production and analysis activities. This LIMS
controls the biological material stocks (clone plates, samples etc), the
rearraying of clones and array printing, as well as the storage of
expression profiling measurements (scanned images and quantification
results). Such an environment is essential for quality control,
traceability, archiving & project data sharing.
Post-Production: expression data analysis is perfomed by exporting data (tabulated text or MAGE-ML) to third party analysis software (J-Express, Cluster etc.)
|
