I’m having a debate with some colleagues about cladistics and character
independence, and I would like to hear the opinion of readers of this group.
I think it's clear that one cannot treat RFLPs as caldistic character data
due to the non-independence of fragment lengths (add one cut site between two
others, and you loose one long fragment and gain two shorter ones).
But what about the "finger-prints" from transposable elements that _contain_
restriction sites? Does tagging just those fragments terminated by the
insertion sequence (mostly at only one end) produce cladistic data? If the
restriction sites within the transposable elements make up only a small
proportion of the total number of restriction sites, then the fragment
lengths produced at different insertion points are largely independent.
My first reaction was that a sub-set of data surely can’t tell you more than
the complete set, but this is not an arbitrary sub-set, it is specifically
the fragment lengths to the left and right of the insertion points of a
transposable element. I argued against it, but I’m now not convinced.
There are several reasons that make finger-prints more desirable than direct
sequencing; it’s easier, cheaper, and there are vastly more finger-prints
already available than good sequences--but only if it produces good cladistic
data.
------------------------------------------------------
Neil Caithness
Epidemiology Research Unit, NCOH, Johannesburg, &
Dept. of Zoology, University of the Witwatersrand
E-mail: neil at gecko.biol.wits.ac.za / neil at is.co.za
