Microarray image gridding with stochastic search based approaches

The paper reports a novel approach for the problem of automatic gridding in Microarray images. Such problem often requires human intervention; therefore, the development of automated procedures is a fundamental issue for large-scale functional genomic experiments involving many microarray images. Our method uses a two-step process. First a regular rectangular grid is superimposed on the image by interpolating a set of guide spots, this is done by solving a non-linear optimization process with a stochastic search producing the best interpolating grid parameterized by a six values vector. Second, the interpolating grid is adapted, with a Markov Chain Monte Carlo method, to local deformations. This is done by modeling the solution a Markov random field with a Gibbs prior possibly containing first order cliques (1-clique). The algorithm is completely automatic and no human intervention is required, it efficiently accounts arbitrary grid rotations, irregularities and various spot sizes.