A New Misalignment-Resistant Algorithm for Microarray Gridding Based on Correlation

Using a various types of microarrays based on multiple biological objects has been grown over recent years and different printing technologies and software solutionsfor extracting data from printed slides have been developed subsequently. Subject to the complexity of printing spots which are made from biological materials, in small scales,there are some printing imperfections including misalignment of spots. One possible way to overcome these problems is using image processing techniques. In this paper, we propose some post-processing algorithms to refine gridding lines which are resulted from autocorrelation-based gridding approaches. The gridline refining algorithm works based on comparison of Pearson Correlation Coefficient(PCC) values of some sequences which are computed from mean or sum of rows and columns in the corresponding directions. Experimental results are shown that the underlying algorithms are successful to correct misalignment of spots and omission of incorrect rows and columns

Keywords:

Miroarray , statisitical pattern recognition , bioinformatics , image proccesing , spot misalignment