> To: molecular-evolution at net.bio.net> From: <jamesmc at acer.gen.tcd.ie>
> Subject: Re:Transition/Transversion=?
> Date: 28 Aug 1995 13:01:15 +0100
> In response to Mark Siddall's posting
> There have been some other postings on this point. But people in glasshouses
> should not throw stones. With more than 8 or 9 taxa it is necessary to use
> a heuristic search procedure to find the most parsimonious tree. Change
> the input order and the tree changes (didn't I just hear that this was a
> problem with NJ?).
I agree whole-heartedly: I have recently had occasion to do trees fro
2 different groups of 30-50 virus sequences - and despite having
PHYLIP and PAUP to run on a mainframe, it got to be too ridiculous a
task to attempt to get the most parsimonious trees out. So we went
to TreeCon 3.0 for Windows, which uses NJ, and bootstraps the data
most effectively. And got tree(s) which agreed very nicely with
trees done using PAUP on subsets of the data. This may not be
entirely kosher in the eyes of the cladisticians; however, I find
that whenever anyone extols the virtues of a philosophy tot he extent
that has been going on here, that they are operating off dogma and
not from logic (or observation).
> >Again I ask how you know when you got the correct tree?????
>> In the context of the original question (which is better etc...) I was giving
> perhaps an oversimplified answer and *correct* may have been too strong a word.
> I am simply reiterating the results of many simulation studies and NJ is said
> to be robust.
Well, there are ways...like taking a sequence, sticking it into a
programme (such as one developed by my colleague Prof Eric Harley,
harley at chempath.uct.ac.za) which mutates it into as many forms as you
want - then taking the output sequences and hitting them with
everything you can throw at them. Neighbour-joining does as good a
job as Hennig and/or Paup. And far quicker than paup and with the
advantage of giving you distances over Hennig.
> Well Mark, if I calculate distance matrices from my data, draw NJ trees,
> bootstrap the data and find the taxa of interest are joined 95% of the time
> and I get the same result using parsimony (i.e. 95%) I am happy and I go for
> a few beers. I don't think you are going to change my mind.
>> Look at the literature...how many scientists publish trees by one method and
> indicate (in the methods section)that they also examined the data by the other
> method and if it agreed then they seem to be happy. The journal editors also
> seem to be happy to publish this.
Too true...I now do everything using Treecon, back-checked on a
subset of the data using PAUP. And I'm happy, and editors have been
too, so far....
| Ed Rybicki, PhD | ed at molbiol.uct.ac.za |
| Dept Microbiology | University of Cape Town |
| Private Bag, Rondebosch | 7700, South Africa |
| fax: x27-21-650 4023 | phone: x27-21-650-3265 |
------WWW URL: http://www.uct.ac.za/microbiology------
"And then one day you find, ten years have got behind you"