Formal concept analysis for the identification of combinatorial biomarkers in breast cancer

  • Authors:

  • Susanne Motameny;Beatrix Versmold;Rita Schmutzler

  • Affiliations:

  • ZAIK, Center for Applied Computer Science, University of Cologne, Germany;Division of Molecular Gyneco-Oncology, Department for Obstetrics and Gynecology, University of Cologne, Germany;Division of Molecular Gyneco-Oncology, Department for Obstetrics and Gynecology, University of Cologne, Germany

  • Venue:
  • ICFCA'08 Proceedings of the 6th international conference on Formal concept analysis
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

When cancer breaks out, central processes in the cell are disturbed. These disturbances are often due to abnormalities in gene expression. The microarray technology allows to monitor the expression of thousands of genes in human cells simultaneously. It is common knowledge that tumor cells show different gene expression profiles compared to normal tissue but also to tissue obtained from metastases. However, the identification of biomarkers, that is sets of genes whose expression change is highly correlated with the disease, poses a great challenge. Increasingly important is the extraction of combinatorial biomarkers. Here, the correlation to the disease is a result of the joint expression of several genes, whereas the single genes do not necessarily distinguish well between healthy and diseased tissue types. In this paper we describe how formal concept analysis can be used to identify gene combinations that are able to distinguish between tumor- and metastasis tissue in breast cancer based on microarray gene expression data.