CrossHybDetector: Detection of cross-hybridization events in DNA microarray experiments

Abstract

Background: DNA microarrays contain thousands of different probe sequences represented on their surface. These are designed in such a way that potential cross-hybridization reactions with non-target sequences are minimized. However, given the large number of probes, the occurrence of cross hybridization events cannot be excluded. This problem can dramatically affect the data quality and cause false positive/false negative results. Results: CrossHybDetector is a software package aimed at the identification of cross-hybridization events occurred during individual array hybridization, by using the probe sequences and the array intensity values. As output, the software provides the user with a list of array spots potentially 'corrupted' and their associated p-values calculated by Monte Carlo simulations. Graphical plots are also generated, which provide a visual and global overview of the quality of the microarray experiment with respect to cross-hybridization issues. Conclusion: CrossHybDetector is implemented as a package for the statistical computing environment R and is freely available under the LGPL license within the CRAN project. © 2008 Uva and de Rinaldis; licensee BioMed Central Ltd.