Dear Mol-evol members:
I am interested in sampling opinions on a subject of practical concern in my
work, which involves epidemiological tracking of pathogenic bacteria with
genetic markers. My main question is, Which method is more likely to be
reliable for reconstructing phylogenetic relationships:
1. RFLP patterns generated using pulsed-field gel electrophoresis (PFGE) of
large DNA fragments generated by rarely-cutting enzymes, OR
2. RFLP patterns generated by Southern hybridization to genomic DNA
restricted by a frequently-cutting enzyme, using an insertion element as a
probe.
Examples of the former approach can be found easily by picking up almost any
current issue of Journal of Clinical Microbiology. The latter approach is
widely used to "type" Mycobacterium tuberculosis, and two seminal papers on
the use of insertion elements to reconstruct phylogenetic relationships
among closely related strains of E. coli are those by Lawrence et al., Mol
Biol Evol 6:1-14 (1989) and Naas T et al., Genetics 136:721-730 (1994).
The main questions I have not been able to resolve for myself concern not
the basic applicability of the methods in a particular instance (e.g., one
can assume that IS elements are present and mobile enough to differn in
insertion-site patterns between strain of interest, and that
high-molecular-weight DNA of the size required for PFGE can be prepared and
appropriately cleaved). Rather, I am most curious about the (hopefully
testable) genetic and evolutionary assumptions that are required to
optimally reconstruct phylogenetic relationships from (1) the behavior of
RFLPs for which the basis is not understood (PFGE) and (2) the transposition
behavior of a mobile DNA element (IS typing).
Thanks in advance for any advice or opinion. Please feel free to respond to
the group.
Cheers,
Mike Coulthart
============================================================
Michael B. Coulthart, Ph.D.
Laboratory Centre for Disease Control
Room 240, HPB Building #7, AL 0701D1
Tunney's Pasture
Ottawa, Ontario
Canada K1A 0L2
Telephone: (613) 952-7312
Facsimile: (613) 941-2408
e-mail: mcoulthart at cyberus.camike_coulthart at hc-sc.gc.ca
