The improvement of the parallel algorithm for randomization-based enrichment analysis

Abstract

This work is motivated by algorithm that is used to analyze the multidimensional genetic data. This algorithm, called permutation test, is widely used as part of gene set enrichment analysis method. In this paper, the permutation test algorithm is considered. The dependence of the resampling test algorithm performance on the input data is studied. Several ways to improve the permutation test algorithm are proposed. The most effective way consisting in the reduction of number of iterations of the algorithm is implemented.