[Taxacom] SINES, LINES and chromosomal rearrangements
Pierre Deleporte
pierre.deleporte at univ-rennes1.fr
Mon May 9 07:35:27 CDT 2011
With permission, John,
may I recall you that I explained you on this list (some years ago now)
that what you call "cladistic" analysis IS clique analysis properly
(=finding the largest set of mutually compatible characters =
"compatibility" analysis, = you select a priori a set of non-homoplastic
characters like Dick Jensen just explained it) -
so I invited your readers to understand "clique analysis" every time you
had written "cladistic";
I also invited you to write "clique analysis" any time you mean it, to
be understood by your readers -
without such a correction, many of your statements concerning
"cladistics" are not (logically) understandable
I observe that you have persisted in using "cladistics" instead of
compatibility-clique analysis
I otherwise suggested that you should use the clique analysis algorithm
to analyse your morphological data,
so that you can realise that this algorithm does exactly what you do by
hand;
and also that you should analyse some molecular data the same way,
finally demonstrating by yourself that molecular data are not "phenetic"
in themselves,
for the very good reason that you would have yourself analysed these
data your proper "cladistic" way (=clique analysis).
last time I exposed this to you, you concluded that you would'nt analyse
molecular data yourself anyway -
now, clique analysis being practically no more in use in contemporaneous
phylogenetics since decades, there is little chance that you ever find a
molecular analysis (or even a morphological analysis) fitting your taste
in the contemporaneous literature
let's see what happens this time... will you perform a molecular clique
analysis and have the sudden revelation that molecular data are not
"phenetic" in themselves?... who bets?
Pierre
Le 06/05/2011 15:28, John Grehan a écrit :
> Perhaps it's my ignorance, but picking the largest set of mutually
> congruent characters seems like it's effectively the same thing as a
> 'tree'. But then I admit my ignorance of the clique analysis algorithm
> and accept I could be wrong in that assertion.
>
> John Grehan
>
>
> -----Original Message-----
> From: taxacom-bounces at mailman.nhm.ku.edu
> [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Richard Jensen
> Sent: Friday, May 06, 2011 9:23 AM
> To: taxacom at mailman.nhm.ku.edu
> Subject: Re: [Taxacom] SINES, LINES and chromosomal rearrangements
>
> Ah, but that was the beauty of clique analysis. One could, without
> reference to any tree, find the greatest number of characters in the
> data set that were mutually congruent. Then one could build a tree on
> which those characters (the maximum clique) illustrated no homoplasy.
> All other characters were, by definition, homoplasious on that tree!
>
> Dick J
>
> On 5/6/2011 9:17 AM, John Grehan wrote:
>> Ok - I get that. As long as there is character incongruence there is
>> homoplasy, although if one is limited to two characters then one
> cannot
>> make a choice as to which. That still appears to correspond to my
> point
>> that one has to have a tree to identify homoplasious features (and
> also
>> determine how much of a 'problem' the level of homoplasy depending on
>> the strength (as measured by whatever indices) of that tree.
>>
>> John Grehan
>>
>> -----Original Message-----
>> From: taxacom-bounces at mailman.nhm.ku.edu
>> [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Richard
> Jensen
>> Sent: Friday, May 06, 2011 9:08 AM
>> To: taxacom at mailman.nhm.ku.edu
>> Subject: Re: [Taxacom] SINES, LINES and chromosomal rearrangements
>>
>> Meacham showed that one did not need a tree to determine the presence
> of
>> homoplasy. The question was, could these two characters support the
>> same tree? He domeonstrated that one could answer that question
> without
>> reference to a tree.
>>
>> This seemed to me a valuable insight, especially if one had reason to
>> hypothesize that one character was likely to be more informative than
>> the other.
>>
>> Dick J
>>
>> On 5/6/2011 8:55 AM, John Grehan wrote:
>>> I think you are just saying what I said in a different way - that
>>> homoplasy is recognized only in relation to the product of analysis -
>>> i.e. the tree. One has to have a result (a tree) to assess whether
>>> individual characters correspond. Those that do not are called
>>> homoplasies (unless I got the totally wrong).
>>>
>>> John Grehan
>>>
>>> -----Original Message-----
>>> From: taxacom-bounces at mailman.nhm.ku.edu
>>> [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Richard
>> Jensen
>>> Sent: Friday, May 06, 2011 8:49 AM
>>> To: taxacom at mailman.nhm.ku.edu
>>> Subject: Re: [Taxacom] SINES, LINES and chromosomal rearrangements
>>>
>>> Sorry, John, but that's not quite right. As Meacham demonstrated in
>> the
>>> early '80's, one can determine, a priori, whether or not any two
>>> characters will support the same tree. Thus, if they disagree, at
>> least
>>> one of them will be homoplasious on any tree. My suspicion is that
>> this
>>> has been much ignored because it was presented in the context of
>>> conducting clique analyses. But the idea extends to any data matrix.
>>>
>>> Dick J
>>>
>>>
>>> On 5/6/2011 8:09 AM, John Grehan wrote:
>>>> The homoplasy argument is ad hoc. The only way own knows about
>>> homoplasy
>>>> is after the analysis. In the case of the human-great ape
>>>> morphogenetics, homoplasy seems to be relatively unproblematic since
>>>> there is such a great majority of congruent characters supporting
> the
>>>> human-orangutan clade.
>>>>
>>>> John Grehan
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: taxacom-bounces at mailman.nhm.ku.edu
>>>> [mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Kenneth
>>> Kinman
>>>> Sent: Wednesday, April 06, 2011 11:48 PM
>>>> To: taxacom at mailman.nhm.ku.edu
>>>> Subject: [Taxacom] SINES,LINES and chromosomal rearrangements (was:
>>>> contamination)
>>>>
>>>> Hi John,
>>>> I'll certainly agree with your second point, that
> examining
>>> the
>>>> human-great ape question will be heuristically valuable and the
>>> results
>>>> could well surprise many researchers long convinced of an exclusive
>>>> chimp-hominid clade.
>>>> However, although we agree that an exclusive
> "chimp-gorilla"
>>> clade
>>>> is probably a very good bet, I would also bet that your contention
>>> that
>>>> there is also an exclusive "orangutan-hominid" clade is due to your
>>>> inordinate distrust of molecular data (even SINEs and LINEs and
> wider
>>>> chromosomal arrangements, which are clearly more reliable than
> simple
>>>> indels, and undoubtedly often more reliable than a lot of
> macroscopic
>>>> morphologies that you seem to prefer). Most of your arguments have
>>> been
>>>> directed at indels (which are clearly less reliable).
>>>>
>>>> Anway, this is another middle ground approach that I
> think
>> is
>>>> superior. Large molecular sequences (SINES, the even larger LINES,
>>> and
>>>> the even larger chromosomal rearrangements) are where we should
>>>> concentrate. Simplistic indels and morphological characters are
>> often
>>>> too subject to homoplasy, so many on both sides of that debate
>>>> ultimately miss the mark.
>>>> -----------Ken Kinman
>>>> P.S. This all sort of reminds me of the political debate in
>>> Washington
>>>> whether the Democrats or Republicans have the best ideas. Those few
>>>> Independents who want to follow a middle ground approach get almost
>> no
>>>> media coverage whatsoever. The media just covers the extremes
>> because
>>>> confrontation "sells" ot the masses in general (sort of like
> sports).
>>>> But the scientific media is even more negligent, because it tends to
>>>> favor one extreme only (strict cladism) and usually ignores any
>>>> arguments for paraphyly (be they moderate or extreme). But I still
>>>> think that many strict cladists will eventually be maligned like
> Wall
>>>> Streeters are increasing being maligned for their extreme (and often
>>>> misleading) views on what would actually benefit the vast majority
> in
>>>> the long run. But thanks to the U.S. government, many Wall
> Streeters
>>>> are still raking in obscene profits playing computer games with
> other
>>>> people's money. Not that strict cladists (or any other biologists)
>>> are
>>>> sharing in that excess of governmental welfare (and lack of
>>> regulation),
>>>> but strict cladists certainly seem to fare a lot better than those
>> who
>>>> are not strict cladists. Not because they are right, but because
>> they
>>>> have more clever lobbyists.
>>>> -------------------------------------------------------------
>>>>
>>>> John Grehan responded to my post:
>>>>
>>>>> perhaps the best bet (my hypothesis) is that an orangutan
>>>>> clade split off next, then a hominid clade, and finally>the
>>>>> chimp-gorilla clade. This would explain the morphological
>>>> similarities
>>>>> of orangtans and hominids (great ape "plesiomorphies" of>two
>> adjacent
>>>>> basal clades).
>>>> It would explain them that way only if that were the case. While Ken
>>> may
>>>> think it's the best bet, in my differing opinion there is no
>> necessary
>>>> imperative for the evidence to lead to that.
>>>>
>>>>> If a lot more attention were paid to proving or disproving>an
>>>>> exclusive chimp-gorilla clade, we might actually get>somewhere, As
>>>> long
>>>>> as the chimp-hominid clade is regarded as "solid", there>will be
>>>>> researchers out there challenging that hypothesis if it is>not
>> backed
>>>> up
>>>>> by strong morphological characters. I agree with John on>that, but
>> I
>>>>> still doubt that this means orangutans and hominids form>an
>> exclusive
>>>>> clade.
>>>> I see the human-great ape question as a heuristic for the challenge
>> of
>>>> dealing with incongruent sequence and morphogenetic evidence.
> Because
>>>> the question of human origins is so prominent the matter is not so
>>> easy
>>>> to sweep under the carpet as it may be with more obscure groups.
>>>>
>>>>
>>>> _______________________________________________
>>>>
>>>> Taxacom Mailing List
>>>> Taxacom at mailman.nhm.ku.edu
>>>> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
>>>>
>>>> The Taxacom archive going back to 1992 may be searched with either
> of
>>>> these methods:
>>>>
>>>> (1) http://taxacom.markmail.org
>>>>
>>>> Or (2) a Google search specified as:
>>>> site:mailman.nhm.ku.edu/pipermail/taxacom your search terms here
>>>>
>>>> _______________________________________________
>>>>
>>>> Taxacom Mailing List
>>>> Taxacom at mailman.nhm.ku.edu
>>>> http://mailman.nhm.ku.edu/mailman/listinfo/taxacom
>>>>
>>>> The Taxacom archive going back to 1992 may be searched with either
> of
>>> these methods:
>>>> (1) by visiting http://taxacom.markmail.org
>>>>
>>>> (2) a Google search specified as:
>>> site:mailman.nhm.ku.edu/pipermail/taxacom your search terms here
>>>
--
Pierre DELEPORTE
UMR6552 EthoS
Université Rennes 1
CNRS
Station Biologique
35380 PAIMPONT
tél (+33) 02 99 61 81 63
fax (+33) 02 99 61 81 88
More information about the Taxacom
mailing list