Biomarker identification by knowledge-driven multilevel ICA and motif analysis

Int J Data Min Bioinform. 2009;3(4):365-81. doi: 10.1504/ijdmb.2009.029201.

Abstract

Traditional statistical methods often fail to identify biologically meaningful biomarkers from expression data alone. In this paper, we develop a novel strategy, namely knowledge-driven multi-level Independent Component Analysis (ICA), to infer regulatory signals and identify biomarkers based on clustering results and partial prior knowledge. A statistical test is designed to evaluate significance of transcription factor enrichment for extracted gene set based on motif information. The experimental results on an Rsf-1 (HBXAP) induced microarray data set show that our method can successfully extract biologically meaningful biomarkers related to ovarian cancer compared to other gene selection methods with or without prior knowledge.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

  • Biomarkers, Tumor / analysis*
  • Biomarkers, Tumor / chemistry
  • Cluster Analysis
  • Computational Biology / methods*
  • Oligonucleotide Array Sequence Analysis
  • Principal Component Analysis

Substances

  • Biomarkers, Tumor