In silico microdissection of microarray data from heterogeneous cell populations

We propose a computational framework for removing the effects of sample heterogeneity by "microdissecting" microarray data in silico. The computational method provides estimates of the expression values of the pure (non-heterogeneous) cell samples. The inversion of the sample heterogeneity can be facilitated by providing accurate estimates of the mixing percentages of different cell types in each measurement. For those cases where no such information is available, we develop an optimization-based method for joint estimation of the mixing percentages and the expression values of the pure cell samples. We also consider the problem of selecting the correct number of cell types.The efficiency of the proposed methods is illustrated by applying them to a carefully controlled cDNA microarray data obtained from heterogeneous samples. The results demonstrate that the methods are capable of reconstructing both the sample and cell type specific expression values from heterogeneous mixtures and that the mixing percentages of different cell types can also be estimated. Furthermore, a general purpose model selection method can be used to select the correct number of cell types.Recent developments in high-throughput genomic technologies have revolutionized the approaches aimed at understanding biological systems and emphasized the need for computational and systems biology research. Microarray analysis, for instance, can provide massive amounts of information about a biological sample by simultaneously measuring thousands of transcript levels. Application of such methodologies has already yielded important molecular insight into cellular phenotypes under various experimental conditions [1] and provided new knowledge about the development and treatment of human diseases, such as cancers [2-4]. During the last several years, microarray technology has undergone continued improvement with better quality control in the overall measurement process, ranging from hybridization conditions