bc-GenExMiner: an easy-to-use online platform for gene prognostic analyses in breast cancer

Breast Cancer Res Treat. 2012 Feb;131(3):765-75. doi: 10.1007/s10549-011-1457-7. Epub 2011 Mar 31.


Gene prognostic meta-analyses should benefit from breast tumour genomic data obtained during the last decade. The aim was to develop a user-friendly, web-based application, based on DNA microarrays results, called "breast cancer Gene-Expression Miner" (bc-GenExMiner) to improve gene prognostic analysis performance by using the same bioinformatics process. bc-GenExMiner was developed as a web-based tool including a MySQL relational database. Survival analyses are performed with R statistical software and packages. Molecular subtyping was performed by means of three single sample predictors (SSPs) and three subtype clustering models (SCMs). Twenty-one public data sets have been included. Among the 3,414 recovered breast cancer patients, 1,209 experienced a pejorative event. Molecular subtyping by means of three SSPs and three SCMs was performed for 3,063 patients. Furthermore, three robust lists of stable subtyped patients were built to maximize reliability of molecular assignment. Gene prognostic analyses are done by means of univariate Cox proportional hazards model and may be conducted on cohorts split by nodal (N), oestrogen receptor (ER), or molecular subtype status. To evaluate independent prognostic impact of genes relative to Nottingham Prognostic Index and Adjuvant! Online, adjusted Cox proportional hazards models are performed. bc-GenExMiner allows researchers without specific computation skills to easily and quickly evaluate the in vivo prognostic role of genes in breast cancer by means of Cox proportional hazards model on large pooled cohorts, which may be split according to different prognostic parameters: N, ER, and molecular subtype. Prognostic analyses by molecular subtype may also be performed in three robust molecular subtype classifications.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Breast Neoplasms / genetics*
  • Breast Neoplasms / mortality
  • Breast Neoplasms / pathology
  • Cluster Analysis
  • Data Mining / methods
  • Female
  • Gene Expression Profiling / methods*
  • Gene Expression Regulation, Neoplastic
  • Humans
  • Internet
  • Middle Aged
  • Prognosis
  • Software*
  • Survival Analysis
  • Young Adult