Comparison of image-transformation methods used in matching 2D gel electrophoresis images

Automatic alignment (matching) of two-dimensional gel electro-phoresis images is of primary interest in the field of proteomics. The method of 2D gel image matching proposed in a previous paper [1] is based on fuzzy alignment of features extracted from gels' images and enables both global and local interpolation of the image grid, followed by brightness interpolation. There are two types of pixel coordinate mapping (forward and inverse) and the choice of mapping method might affect the quality of the results obtained. In this paper the two types of grid mapping method are compared.