The LOD scores between every pair of markers, derived in the analysis, is used to estimate the distance between each pair

Based upon these distances, markers are placed within multi-dimensional space using an NxN matrix of inter-marker distances. Where possible, distally positioned links (which are inherently inaccurate) are replaced by several shorter more accurate distances.

At each iteration, the weakest link is eliminated and using distance geometry, the complexity reduced by one dimension.

Eventually a three dimensional representation of the map is reached (click the 3d image to rotate). This can simply be translated perpendicular to the principal axis to produce a linear map, or estimated errors can be used to position markers according to confidence.