Chaos automata: iterated function systems with memory

Transforming biological sequences into fractals in order to visualize them is a long standing technique, in the form of the traditional four-cornered chaos game. In this paper we give a generalization of the standard chaos game visualization for DNA sequences. It incorporates iterated function syste...

Full description

Saved in:
Bibliographic Details
Published in:Physica. D 2003-08, Vol.181 (3), p.274-285
Main Authors: Ashlock, Dan, Golden, Jim
Format: Article
Language:English
Subjects:
Biological fractals
Chaos games
Evolutionary computation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Transforming biological sequences into fractals in order to visualize them is a long standing technique, in the form of the traditional four-cornered chaos game. In this paper we give a generalization of the standard chaos game visualization for DNA sequences. It incorporates iterated function systems that are called under the control of a finite state automaton, yielding a DNA to fractal transformation system with memory. We term these fractal visualizers chaos automata. The use of memory enables association of widely separated sequence events in the drawing of the fractal, finessing the “forgetfulness” of other fractal visualization methods. We use a genetic algorithm to train chaos automata to distinguish introns and exons in Zea mays (corn). A substantial issue treated here is the creation of a fitness function that leads to good visual separation of distinct data types.
ISSN:0167-2789
1872-8022
DOI:10.1016/S0167-2789(03)00122-2