Development and optimization of experiments in biology laboratories requires flexible solutions to respond to requirements concerning

v      data heterogeneity,
v      the combination of data obtained from distributed public external databases (EMBL, Gene Ontology, SwissProt, PubMed…) and of private data,
v      the access by biologists to the analytic process,
v      complete traceability of data management and analysis processes,
v      easy diffusion of experiments and the provision of results.

Based on such innovative technologies as Data Morphing and on specific data processing, Amadea BioPack is an answer to these needs. It provides biologists and doctors with an easy-to-use homogeneous environment and immediate access to public databases and biological data worldwide. Amadea BioPack is the answer in any domain of biological research confronting these difficulties. For instance, the fight against cancer benefits from the rapid development of tumor analysis techniques. Amadea BioPack is used in many laboratories to resolve the problem of the integration of biological and anatomo-clinical data obtained from analysis of a tumor or from a group of tumors.

Data access

Amadea BioPack provides, without programming, a direct connection to data, distant or local: reference databases for the biological community and local data of the laboratory. This data can contain information about genes, chromosomes, sequences, patients or any other type of information from clinical research, microarray analyses (CGH, Affymetrix, …), or external databases for the biological community (EMBL, Gene Ontology, Refseq, Pubmed,…).

Data linking

Thanks to its generic toolbox that allows biological analyses to be extended without programming, Amadea BioPack considerably reduces the time needed to integrate new data sources and external processing.

At any step of the analysis, external information can be linked in real-time and transparently to local laboratory data thanks to predefined operators. For instance, it is possible to link information concerning sequences, genes, ETSs, proteins, diseases, metabolic functions, cellular components, biological processes, publications, etc. so as to enrich experimental data and highlight new information.

Data processing automation

Fig. 1: Class analysis – Web-based analysis of the Gene Ontology terms best differentiated between two groups of transcriptome experiments.

Amadea BioPack supports, interactively and without programming, a data-driven experimentation methodology. The users have access to: Parameterizable functional blocks for specific processing, for example normalization and filtering of microarray data. The possibility of joint analysis of microarrays built with different technologies, and enriching the results with external information. The capacity to integrate external processes (scripts, algorithms or tools) for an optimal use of existing resources to give a single application platform, thus guaranteeing an optimal traceability. The diffusion of results and the immediate generation of web interfaces to pilot the analyses.

Performances

Amadea BioPack greatly reduces the time needed for the development phase of an analysis. Typically, substantial time is saved (5- to 30-fold) in comparison to classical programming languages (SQL, PERL or others).

Based on Data Morphing, a technology developed by ISoft, Amadea BioPack is designed to process very large volumes of data (hundreds of millions of rows) very quickly. It has its own cache memory management mechanism, which optimizes the use of the machine memory according to available resources. The software easily adapts to available resources, thus providing a simple and evolutive solution.

For instance, retrieving all ESTs for which 3' and 5' extremities are available in public databases, will identify 500 000 sequences from the 5 000 000 EMBL entries, and is expressed in Amadea BioPack with 5 operators, requires only a few minutes to be developed and 5 minutes to execute on a PC.