Sub-grid detection in DNA microarray images

Abstract

Analysis of DNA microarray images is a crucial step in gene expression analysis, as it influences the whole process for obtaining biological conclusions. When processing the underlying images, accurately separating the subgrids is of supreme importance for subsequent steps. A method for separating the sub-grids is proposed, which aims to first, detect rotations in the images independently for the x and y axes, corrected by an affine transformation, and second, separate the corresponding sub-grids in the corrected image. Extensive experiments performed in various real-life microarray images from different sources show that the proposed method effectively detects and corrects the underlying rotations and accurately finds the sub-grid separations. © Springer-Verlag Berlin Heidelberg 2007.