DNA Sequence Analysis Using Hierarchical ART-based Classification Network

Adaptive resonance theory (ART)describes a class of artificial neural networkarchitectures that act as classification tools whichself-organize, workin realtime, and require no retraining to classify novel sequences. Wehave adapted ARTnetworks to provide support to scientists attempting to categorize tandemrepeat DNAfragments from Onchocerca volvulus. In this approach, sequences of DNAfragments are presented to multiple ART-basednetworks which are linked together into two(or more)tiers; the first provides coarse sequenceclassification while the subsequenttiers refine the classifications as needed.The overall rating of the resulting classification of fragmentsis measuredusing statistical techniques based on those introduced by Zimmerman,et al. (1994) validate results from traAitional phylogeneticanalysis. Tests of the Hierarchical ART-basedClassification Network, or HABclassnetwork, indicate its value as a fast, easy-to-useclassification tool whichadaptsto newdata without retraining on ...

• Type: Conference
• Year: 1994
• Where: ISMB
• Authors: Cathie LeBlanc, Charles R. Katholi, Thomas R. Unnasch, Susan I. Hruska
• Keywords: Adaptive Resonance Theory, Artificial Neural Networkarchitectures, Computational Biology, ISMB 1994, Onchocerca Volvulus