A systematic approach for outlier detection and its correction on oligonucleotide microarray

摘要

The advent of microarray technologies gives an opportunity to monitor the expression of ten thousands of genes, simultaneously. Such microarray data can be deteriorated by experimental errors and image artifacts, which generate non-negligible outliers that are estimated by 1 5% of typical microariay data. Thus, it is an important issue to detect and correct these faulty probes prior to high level data analysis such as classification or clustering. In this paper, we propose a systematic procedure for the detection of faulty probes and its proper correction in Affymetrix Genechip array based on multivariate statistical approaches Principal component analysis, one of the most widely used multivariate statistical approaches, has been applied to construct a statistical correlation model with 20 pairs of probes for each gene. And, the faulty probes are identified by inspecting the squared prediction error (SPE) of each probe. Then, the outlying probes are reconstructed by the iterative optimization approach minimizing SPE. Through the application study, the proposed approach showed good performance for probe correction without removing faulty probes, which may be desirable in the viewpoint of the maximum use of data information.