1. Hauptnavigation
  2. Navigation des Hauptbereiches
  3. Inhalt der Seite

FACT





For general users:

FACT can supply annotation data for your experimental results.

FACT can help you discover patterns in your annotated data.

FACT alliviates the interpretation of the results of high-throughput experiments like microarrays

For developers:

FACT is a project which grows through the addition of new annotation sources or different analysis functions.
Additions can be easily made by writing and uploading a module, using the supplied prototype.
Contact us, if you have new sources, ideas, or modules: support@factweb.de

General introduction:

High-throughput experiments like DNA-Microarrays generate extremely large amounts of data. A variety of algorithms and programs have been introduced to accomplish the processing of raw data as well as the statistical analysis of the results. But besides the mathematical complexity that needs to be handled, there is a biological complexity inherent to the data sets, too. Computational means to acquire an interpretation of the biologically complex data have only been addressed insufficiently. Nevertheless the results are often too complex to be efficiently handled by humans. Common questions in the interpretation of results can be generalized as the following.


  • Are there correlations with other experimental results (experimental data)?
  • What information (annotation data) is known about the analyzed features (genes, e.g.)?
  • Are there correlations between the experimental outcomes and the additional information (analysis data)?


© dkfz.de



The program Flexible Annotation And Correlation Tool (FACT) is being developed to address these questions. It is set up on top of a database that reflects the structure of the data as experimental, annotation and analysis data. A core software framework written in the language PERL supplies the base-functions for adding and operating modules of the three categories:

  • Different types of experimental data can be loaded using different parser functions (e.g. microarray data).
  • Varying sources of annotations can be acquired for the experimental data by different data-access functions (e.g. web-access to the ensemble database).
  • Correlation and analysis schemes can be used depending on the loaded correlation functions (e.g. GO-term analysis).


Vergrößerte Ansicht GeneOntology analysis of the molst relevant terms. Terms from the Function Ontology exceeding a pvalue cutoff of 0.009 . | © dkfz.de

The main focus of FACT is to alleviate the task of analyzing biological complexity by providing a concise framework that represents the basic structure of many analysis procedures used by biologists today. Versatility of the tool is accomplished by providing a modular structure that allows plugging in new features on a need-to-need basis with the main program as a perl library, which could also be employed by other analysis tools.





last update: 26/02/2010 back to top