distance based DNA trees

[B.J.Tindall]

In direct reply to Herb and Jason I deliberately include "parsimony of
evolution" in quotation marks, because, yes there is "parsimony of logic"
and also parsimiony operating in evolution, which are certainly two
slightly different things. In the case of maximum likelihood one could, of
course take a path which is not the most probable, but don't we tend to
look for the most probable solution (most likely answer) or something close
to it, unless we have evidence against it - in which case it isn't the
"most probable solution"? Isn't there a logical link between probability
and "parsimony of logic"? If I recall Felsenstein also indicated that the
likliehood of a particular topology being found was not directly equivalent
to that topology being the "right" answer.
As for maximum parsimony, my question would be how do you arrive at the
"best phylogenetic hypothesis"? I would certainly agree that using the
minimum number of ad hoc assumptions is advantageous, but is this also
something which is specific for maximum parsimony?
Brian

>ALL methods of evaluation,
> whether it be neighbor-joining, maximum parsimony, or even maximum
> likelihood include an underlying assumption that "parsimony of evolution"
> is operating. Thus both phenetic and cladistic approaches include some

Maybe that depends on what you mean by 'parsimony of evolution'.  In maximum
likelihood phylogenetics, all possible paths between observed states
(weighted by their likelihood) are considered, not just the most probable
one.  What is the underlying parsimony there?

> underlying reference to parsimony, without which just anything would be
> possible.

The probabilist view is indeed that anything is possible, though not all
things are probable.  Hence, the probabilistic approach is to calculate the
likelihood of observed character states as the sum of all possible ways of
getting there...

Cheers,
- Jason



At 16:15 15.4.2004 EDT, HJJACOBSON at AOL.COM wrote:
>In a message dated 4/15/2004 12:19:06 AM Pacific Standard Time, bti at DSMZ.DE
>writes:
>
>> Interesting comments, which can be explained. ALL methods of evaluation,
>> whether it be neighbor-joining, maximum parsimony, or even maximum
>> likelihood include an underlying assumption that "parsimony of evolution"
>> is operating. Thus both phenetic and cladistic approaches include some
>> underlying reference to parsimony, without which just anything would be
>> possible.
>> Brian
>
>In think Maximum Parsimony only assumed that the best phylogentic hypothesis
>was the one with the least number of ad hoc assumptions. The parsimony was in
>the logic, not in the "parsimony of evolution." At least it did in the
>beginning. Has it changed?
>
>Herb
>


********************************************************************
* Dr.B.J.Tindall      E-MAIL bti at dsmz.de                           *
* DSMZ-Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH *
* Mascheroder Weg 1b, D-38124 Braunschweig, Germany                *
* Tel.: ++ 531 2616 0 (general)                                    *
* Tel.: ++ 531 2616 224 (direct)                                   *
* Fax:  ++ 531 2616 418                                            *
*                                                                  *
* Homepage: http://www.dsmz.de/index.html                          *
* E-MAIL: contact at dsmz.de (general enquiries)                      *
*         sales at dsmz.de (sales)                                    *
********************************************************************