Skip to main content
. Author manuscript; available in PMC: 2012 Jan 13.
Published in final edited form as: Proteomics. 2010 Dec;10(23):4226–4257. doi: 10.1002/pmic.200900635

Figure 3.

Figure 3

The first four scales of multi-resolution image-based 2-D gel alignment, as illustrated with the RAIN algorithm [68, 70]. (a) Two overlaid gels, one in magenta, one in green, showing the range of geometric deformations and intensity inhomogeneities between them. (b) The top row shows the multi-resolution pyramid for the two gels, with variance-stabilised pixel intensities. The middle and bottom rows show respectively the regionally varying multiplicative and addition spatial bias between the two gels, as modelled with hierarchical piecewise cubic B-splines. (c) The first four scales of alignment with RAIN (there are 7 in total). At each scale, finer and finer deformations are accounted for with a hierarchical piecewise cubic B-spline transformation. Elements reproduced from [68] with author permission under the Creative Commons Attribution-Non-Commercial license.