Incremental multiple sequence alignment

Abstract

This work proposes a new approach to the alignment of multiple sequences. We take profit from some results on Grammatical Inference that allow us to build iteratively an abstract machine that considers in each inference step an increasing amount of sequences. The obtained machine compile the common features of the sequences, and can be used to align these sequences. This method improves the time complexity of current approaches. The experimentation carried out compare the performance of our method and previous alignment methods. © Springer-Verlag Berlin Heidelberg 2007.