In silico identification of gene expression meta-signature that predicts breast cancer prognosis.

LACUNZA E; BUTTI M; ABBA MC

Universidad Nacional de Quilmes, Quilmes, Buenos Aires, Argentina.

Congreso; 1er Congreso Argentino de Bioinformática y Biología Computacional.; 2010

BackgroundGobal gene expression in breast cancer has been profiled extensively over the last decade,providing novel information of biological and clinical relevance for the classification of breastcancers according to specific phenotypes. In this study, we provide a systematic analysis of thegene expression signature derived from 42 breast cancer gene expression studies in an effort toidentify the most relevant breast cancer biomarkers using a novel gene list meta-analysismethod.Materials and methodsThe study approach underwent three phases: (a) detection of overlapping genes among thedifferent signatures, (b) examination of the relationship between gene expression signatures bya two-way unsupervised analysis, followed by (c) identification of the molecular pathways thatare mainly affected by the gene expression meta-signature.ResultsAnalysis of gene expression signatures metadata revealed a set of 117 genes that were themost commonly affected, ranging from 12% to 36% of overlapping among breast cancer geneexpression studies. Data mining analysis of the indicated core of transcripts and protein-proteininteractions of this commonly modulated genes indicate three functional modules significantlyaffected among signatures, one module related to the estrogen receptor alpha and, and twomodules related to the cell cycle signaling pathway, in addition to the targeting of various breastcancer-causing genes. Unsupervised statistical analysis of gene expression metasignaturederived from publicly available data (n=295) identified two main clusters of breast carcinomaswhich differed in their lymph node status (p<0,001), overall survival (p<0,001), relapse-freesurvival (p<0,001), and metastasis-free interval (p<0,001).ConclusionsOur comprehensive comparison of overlapping genes across 42 breast cancer gene expressionsignatures provides an integrated view of a significant number of transcripts identified as highlymodulated in breast tumors. The identification of individual proteins is of high relevance not onlyfor the potential value as prognostic biomarkers but also because may provide insight intomechanisms and pathways of relevance in breast cancer progression. More importantly, thisanalysis identified the most promising biomarkers for further evaluation in breast cancer such asthe cell cycle and mitotic spindle related genes.