Long-range genomic patterns

[Massimo Pigliucci]

     Dear Compunetters,
                         in response to Mike Levin's request
about long-range genomic patterns, there are two types of studies
that I'm aware of (I'm not a molecular biologists, by the way):
the search for long-range correlations in DNA sequences through
algorithms capable of locating fractal (i.e., scale-invariant)
features. A brief review of recent attempts is in Amato, I.
Science 257:747 (1992), but you might want to check several of
the latest issue of Science.
     I'm more familiar with the second approach, because James
Lyons-Weiler and Don Les at the University of Connecticut are
working on it. It encompasses the use of Chaos Game
Representation algorithms (again, for the identification of
fractal patterns) on sequences of the same gene in different
organisms or of related sequences in the same taxon. The first
paper I knew of about this was published by Jeffrey on Nucleic
Acid Research (18:2163-2170, 1990), followed by an interesting
and more detailed work by Hill et al. on the Journal of Molecular
Evolution (35:261-269, 1992).
     Lyons-Weiler has demonstrated convincingly that the fractal
patterns originated by CGR is entirely due to base and
dinucleotide biases, and quite independent from the particular
sequence (it was previously suggested that both the biases and
the sequence might have a part in it). I think it is a great way
to study broad-range constraints, because of the easy
visualization that lends itself to quantification. I doubt the
technique could be of any phylogenetic importance, as instead
suggested by both Jeffrey and Hill et al. This is because, being
the pattern completely independent from the sequence (and so from
the order of input of the data), it does not make any sense to
compare the CGR plots point by point: as in all fractals, what it
counts is the overall image, not the single points that
constitute it.
     Hope I have been of some help. Bye, M.

*****************************************************************
Massimo Pigliucci
Department of Ecology & Evolutionary Biology
University of Connecticut, Storrs CT 06269

phone: 203-486-4372 fax: 203-486-6364
bitnet: pigliucc at uconnvm internet: pigliucc at uconnvm.uconn.edu

"This is becoming really insignificant". "Not enough".
(Samuel Beckett, Waiting for Godot)
*****************************************************************