Advances in Statistical Modeling of High Dimensional Data:
Variable selection, and Challenges in Image Analysis

Markus Schmidberger, LMU München

Estimating Networks in a Huge Microarray Meta-Analysis with 60 Experiments and more than 7000 Microarrays

Public available data sets were collected from public microarray databases and preprocessed and analyzed together. Data from more than 60 experiments, more than 7000 microarrays and eight different cancer entities were used to demonstrate the power of the parallel computing with the 'affyPara' package, to discuss the difficulties of data management, and to estimate gene interaction networks with the 'pcalg' package . This is one of the first projects for analyzing this amount of public available data sets together. Therefore this talk presents more technical details and problems of performing this kind of analyses. The comparison of gene networks for different pathways and different cancer entities partly confirms established forms of gene-gene interaction.

[Back to the Schedule]