CMC : A novel clustering method for human sequence classification

We propose a clustering method for sequence classification based on a recently proposed algorithm that provides a hierarchical clustering under very general assumptions, relying on the co-operative behavior of an inhomogeneous lattice of chaotic coupled maps. The proposed method (Chaotic Map Clustering) has low complexity as compared to the most widely spread methods for sequence classification, surely an appealing feature if dealing with large data-sets. The method incorporates a biologically motivated distance measure based on the introduction of a site variability index. Here we report results obtained by applying the method to two distinct data-sets of human mtDNA HVRI haplotypes from different geographical origins. A comparison with performance resulting from the application of well known methods such as the Stochastic Stationary Markov method and the Reduced Median Network has also been produced.