PHYLOGENETIC RECONSTRUCTION USING AN UNSUPERVISED GROWING NEURAL-NETWORK THAT ADOPTS THE TOPOLOGY OF A PHYLOGENETIC TREE

We propose a new type of unsupervised, growing, self-organizing neural network that expands itself by following the taxonomic relationships that exist among the sequences being classified. The binary tree topol ogy of this neutral network, contrary to other more classical neural n etwork topologies, permits an efficient classification of sequences. T he growing nature of this procedure allows to stop it at the desired t axonomic level without the necessity of waiting until a complete phylo genetic tree is produced. This novel approach presents a number of oth er interesting properties, such as a time for convergence which is, ap proximately, a lineal function of the number of sequences. Computer si mulation and a real example show that the algorithm accurately finds t he phylogenetic tree that relates the data. All this makes the neural network presented here an excellent tool for phylogenetic analysis of a large number of sequences.