Molecular Evolution Modeled as a Fractal Renewal Point Process in
Agreement with the Dispersion of Substitutions in Mammalian Genes
David R. Bickel,
1
Bruce J. West
2
1
Houston Health Science Center, University of Texas, 1100 Holcombe No. 4.430, Houston, TX 77030, USA
2
Center for Nonlinear Science, University of North Texas, P.O. Box 5368, Denton, TX 76203-5368, USA
Received: 12 January 1998 / Accepted: 19 May 1998
Abstract. A fractal renewal point process (FRPP) is
used to model molecular evolution in agreement with the
relationship between the variance and the mean numbers
of nonsynonymous and synonymous substitutions in
mammals. Like other episodic models such as the doubly
stochastic Poisson process, this model accounts for the
large variances observed in amino acid substitution rates,
but unlike certain other episodic models, it also accounts
for the increase in the index of dispersion with the mean
number of substitutions in Ohta’s (1995) data. We find
that this correlation is significant for nonsynonymous
substitutions at the 1% level and for synonymous substi-
tutions at the 10% level, even after removing lineage
effects and when using Bulmer’s (1989) unbiased esti-
mator of the index of dispersion. This model is simpler
than most other overdispersed models of evolution in the
sense that it is fully specified by a single interevent prob-
ability distribution. Interpretations in terms of chaotic
dynamics and in terms of chance and selection are dis-
cussed.
Key words: Overdispersion — Episodic evolution —
Fractal renewal process — Nonlinear — Chaos — Neu-
tral theory
Introduction
Based on the study hemoglobin chains, Zuckerkandl and
Pauling (1962) suggested that the number of amino acid
substitutions between homologous chains in different
species is roughly proportional to the time since the di-
vergence of those organisms from a common ancestor. In
comparing the number of cytochrome c amino acid dif-
ferences between different organisms, Margoliash (1963)
noted that the number of differences between members
of a taxon and organisms outside that taxon is constant.
For example, the number of differences between any of
the mammal sequences and the bird sequence is about
equal. He concluded that the number of substitutions
between two organisms is a function of the time that
elapsed since the two species diverged from a common
ancestor. Zuckerkandl and Pauling (1965) coined the
term ‘‘molecular clock’’ to designate constant rates of
molecular evolution. Progress in nucleic acid sequencing
led to the concept of a DNA clock in addition to the
protein clock since synonymous and nonsynonymous
substitutions seem to occur at different rates.
Although the substitution rates in genes appear to be
approximately constant over long periods of time, they
are not precisely constant. The neutral theory claims that
the rate of evolution is stochastically constant, much as
the rate of radioactive decay is stochastically constant;
see Easteal et al. (1995). However, Gillespie (1991, p.
119) has demonstrated that the differences in the number
of substitutions calculated from molecular data are more
pronounced than those expected from a simple Poisson
process, even when lineage effects are removed. He uses
the index of dispersion, I(t), the ratio of the variance in
the number of nucleotide substitutions to the mean num-
ber of substitutions in time t, as a measure of how much
variation a process has. This quantity was first intro-
duced by Fano (1947) to quantify dispersion in ioniza- Correspondence to: D.R. Bickel; e-mail: BickelDR@aol.com
J Mol Evol (1998) 47:551–556
© Springer-Verlag New York Inc. 1998