Next: Multiple alignment - Up: Tree based alignment Previous: Tree based alignment


It is currently not possible to say which approach to multiple alignment gives the best results. The two methods give approximately equivalent alignments when the sequences are quite similar (as judged by significance scores from pairwise comparisons). If the sequences consisted of two distinct families then the tree based alignment will give better alignments WITHIN the families but roughly equivalent alignments BETWEEN the families. In the current implementation the single order method will run faster so is probably the general method of choice.