Multiple sequence alignment algorithms for the phylogenic analysis of chloroplast DNA

Abstract

The application of approximate string matching and alignment algorithms to either DNA or amino acid (protein) sequences is important for determining conserved regions, functional sites and to allow for multiple sequence alignments from which an evolutionary (phylogenic) tree may be inferred. Global sequence alignment algorithms attempt to maximise the alignment score by placing gaps (which are seen as insertion/deletion evolutionary events) in either sequence so as to maximise the number of matching characters and minimise mismatches and gaps. The extension of traditional dynamic programming algorithms for aligning two sequences to aligning N sequences leads to a polynomial increase in the space and time complexity. Consequently many heuristic multiple sequence alignment algorithms, and improvements in the representation of a multiple sequence alignment, have been developed. This honours project has focussed on multiple sequence alignment algorithms, their processing and space requirements and the suitability of the alignments of samples of chloroplast DNA to further phylogenic analysis. Since the samples being used in the analysis are hypervariable, this research has also looked at algorithms capable of handling inversions in the DNA sequences (where a section of DNA has undergone the mutation of reversing).