To overcome the above mentioned difficulties, and to deal with other situations that cause problems for alignment programs, we implemented a semi-automatic anchored alignment approach. Here, the user can specify an arbitrary number of anchoring points in order to guide the alignment procedure. Each anchor point consists of a pair of equal-length segments of two of the input sequences. An anchor point is therefore characterised by five coordinates: the two sequences involved, the starting positions in these sequences and the length of the anchored segments. As a sixth parameter, our method requires a score that determines the priority of an anchor point. The latter parameter is necessary, since it is in general not meaningful to use all anchors proposed by the user. It is possible that the selected anchor points are inconsistent with each other in the sense that they cannot be included in one single multiple output alignment, see [16] for our concept of consistency. Thus, it may be necessary for the algorithm to select a suitable subset of the proposed anchor points.