Array Mining

ArrayMining is a server for automating statistical analysis of gene and protein expression microarray data, designed as a supporting tool for investigation of the genetic components of diseases. It performs five common gene expression analysis tasks:

• cross-study normalisation
• feature selection
• sample clustering
• sample classification
• network analysis
• gene set analysis

Unlike other microarray-related servers, ArrayMining is using ensemble and consensus techniques (e.g. ensemble feature selection, ensemble prediction, consensus clustering) and performs automatic parameter selection. For a given analysis task it is possible to combine multiple algorithms and data sets in a semi-automatic fashion. This way new exploratory routes become available, e.g. ensemble sample classification can be performed with predictors obtained from a gene set analysis applied to combined data from multiple studies.

The analysis is further simplified by the integration with annotation databases. This enables further functional analysis and literature mining. The results are presented as interactive sortable tables and three dimensional VRML visualizations.

Publications

• DOI BibTeX
  E. Glaab, J.M. Garibaldi, N. Krasnogor
  ArrayMining: a modular web-application for microarray analysis combining ensemble and consensus methods with cross-study normalization
  in BMC Bioinformatics, 10(1):358, 2009

  @ARTICLE{Glaab2009,
    title = {ArrayMining: a modular web-application for microarray analysis combining ensemble and consensus methods with cross-study normalization},
    author = {Glaab, Enrico and Garibaldi, Jonathan M and Krasnogor, Natalio},
    year = 2009,
    doi = {10.1186/1471-2105-10-358},
    journal = {BMC Bioinformatics},
    volume = {10},
    number = {1},
    pages = {358}
  }