Parallelized preprocessing algorithms for high-density oligonucleotide arrays

Studies of gene expression using high-density oligonucleotide microarrays have become standard in a variety of biological contexts. The data recorded using the microarray technique are characterized by high levels of noise and bias. These failures have to be removed, therefore preprocessing of raw data has been a research topic of high priority over the past few years. Actual research and computations are limited by the available computer hardware. Furthermore most of the existing preprocessing methods are very time consuming. To solve these problems, the potential of parallel computing should be used. For parallelization on multicomputers, the communication protocol MPI (message passing interface) and the R language will be used. This paper proposes the new R language package affyPara for parallelized preprocessing of high-density oligonucleotide microarray data. Partition of data could be done on arrays and therefore parallelization of algorithms gets intuitive possible. The partition of data and distribution to several nodes solves the main memory problems and accelerates the methods by up to the factor ten.