Multi-platform data integration in microarray analysis

An increasing number of studies have profiled gene expressions in tumor specimens using distinct microarray platforms and analysis techniques. With the accumulating amount of microarray data, one of the most challenging tasks is to develop robust statistical models to integrate their findings. We compare some recent methodologies on the field, with respect to ER status, and focus on a unified among platforms scale suggested by Parmigiani et al.(2002) and Shen et al.(2004), which is based on a Bayesian mixture model. Under this unified scale, we study the intensity similarities between four breast cancer data sets derived from various platforms. We evaluate our results with an independent data set in terms of ER sample clustering given the derived gene ER signatures of the integrated data. We found that intensity and fold-change variability similarities between different platform measurements can greatly assist the statistical analysis of independent microarray data sets.