Data Analysis and Management

Microarray technology produces huge quantities of data, and this is one of its disadvantages (See Problems of DNA Microarrays). In fact, the development of methods that effectively organise and interpret the voluminous quantities of data generated is the most significant challenge for researchers.
It is vital that information systems such as the one used in this technique are flexible. It is necessary that they are able to receive new statistical data mining tools, as they develop.
It is certain that DNA microarrays will drive bioinformatics software forward.

After hybridisation and the readout of the expression levels obtained from the array, the vast quantities of data that is collected has to be stored and saved. This then enables processing and analysis - both statistical and biological - to be carried out on the data. This is vital, because without significant image processing, the information content of a microarray is very limited.
Affymetrix's LIMS Screen

LIMS

"Laboratory Information Management Systems"

In order to fulfil the potential of large-scale functional genomics technologies, powerful systems are needed to follow and manage the data and the flow of information.
This is the purpose of a LIMS, whose general principles include:

The individual elements of a LIMS, as well as the complex design, is specific to particular laboratories, and their work.
The microarray LIMS that has been developed is called ArrayDB.

ArrayDB

  • ArrayDB is a first-generation, flexible and convenient data management and analysis system for microarrays.

  • ArrayDB maintains the association between a spot on the microarray and an image, and all the information concerned with the clone that is located at that position on the microarray.

  • The ArrayDB system integrates the many processes that make up the technique of microarrays:
    • robotic printing
    • array scanning
    • image processing
    • user interface
    • data management

  • The ultimate aim of ArrayDB is the identification of patterns and relationships among the fluorescence ratios - in individual, and also across multiple experiments.

  • ArrayDB has been developed to manage and analyse large-scale expression data. It has been designed to permit versatility in the type and organisation of the input data, and does allow the inputting of data from a variety of sources. The design also allows the addition of newly isolated clones, for which GenBank accession names or numbers are not available. Most of the clone information that is stored in ArrayDB is extracted from UniGene.

  • The system will facilitate the interpretation of complex hybridisation readouts with diverse options for retrieval and analysis of data.

    • Information stored by ArrayDB

    Information about the individual clones in the microarray:     Information about the fabrication of the microarray, and conditions of the experiment:
    • details concerning printing of the arrays:
      • printer robot parameters
      • environmental conditions
        • temperature
        • humidity
        • tip wash conditions
    • GIPO ("Gene In Plate Order") data - this relates the clones, to their specific order in the plates
    Information about the probes, and other experimental conditions, including:
    • name of the investigator
    • purpose of the experiment
    • textual descriptions of the conditions of the original cell or tissue types
    Information about the hybridisation results:
    • scanned images
    • hybridisation intensity data
    • intensity ratios
    • background values


    To Top


    Back to Index


    To Procedure