Home  · Classes  · Annotated Classes  · Modules  · Members  · Namespaces  · Related Pages

Example 4: Map alignment

Goal 1: Peak map alignment algorithms should map a number of peak maps onto comparable retention time and m/z dimensions.

TOPP_map_mapping_peaks.png

Goal 2: On consensus and feature maps, the alignment approaches should find and group corresponding elements in the maps.

TOPP_map_mapping_features.png

The TOPP tool MapAlignment solves both alignment problems. Our approach is based on the combination of pairwise map alignments. A pairwise map alignment proceeds in several steps. In the first step, the retention time warp and the distortion in m/z is estimated using a pose clustering approach. This initial transformation is used to find elements in the two maps which likely belong together. In the second step, these pairs are used as landmarks and a final, improved transformation is estimated by which the two maps are mapped onto each other in a third step. The first three steps constitute the so called superposition phase. In case of a feature map alignment, the corresponding elements are grouped together in a fourth step, the so called consensus phase.

In addition to the MapAlignment, we offer three tools for the superposition phase of a pairwise feature map alignment which are UnlabeledMatcher, MapMatcher and MapDewarper. These tools can be used if you want a fine-grained control over the matching process or if you are not interested in the feature pairs, but in the actual mapping function.

TOPP_map_mapping_modules.png

Most important MapAlignment parameters


Generated Tue Apr 1 15:36:40 2008 -- using doxygen 1.5.4 OpenMS / TOPP 1.1