Fractal Landscapes and Molecular Evolution:
Modeling the Myosin Heavy Chain Gene Family
S. V. Buldyrev, A. L. Goldberger, S. Havlin, C-K. Peng, H.E. Stanley,
M.H.R. Stanley, and S. Simon,
Biophysical Journal, 65, 2673--2679 (1993).
Abstract
Mapping nucleotide sequences onto a "DNA walk" produces a novel representation
of DNA that can then be studied quantitatively using techniques derived from
fractal landscape analysis. We used this method to analyze 11 complete genomic
and cDNA myosin heavy chain (MHC) sequences belonging to 8 different species.
Our analysis suggests an increase in fractal complexity for MHC genes with
evolution with vertebrate > invertebrate > yeast. The increase in complexity
is measured by the presence of long-range power-law correlations, which are
quantified by the scaling exponent alpha. We develop a simple iterative model,
based on known properties of polymeric sequences, that generates long-range
nucleotide correlations from an initially noncorrelated coding region. This
new model-as well as the DNA walk analysis-both support the intron-late
theory of gene evolution.
