DNA Sequence Analysis using

Adaptive resonance theory (ART) describes a class of artiicial neural network architectures that act as classiication tools which self-organize, work in real-time, and require no retraining to classify novel sequences. We have adapted ART networks to provide support to scientists attempting to categorize tandem repeat DNA fragments from Onchocerca volvulus. In this approach, sequences of DNA fragments are presented to multiple ART-based networks which are linked together into two (or more) tiers; the rst provides coarse sequence classiication while the subsequent tiers reene the classiications as needed. The overall rating of the resulting classiication of fragments is measured using statistical techniques based on those introduced by Zimmerman, et al. (1994) to validate results from traditional phylogenetic analysis. Tests of the Hierarchical ART-based Classiication Network, or HABclass network, indicate its value as a fast, easy-to-use classiication tool which adapts to new data without retraining on previously classiied data.