Talk:Molecular systematics

Untitled
We have information about this area scattered around a number of pages relating to evolution, genetics and taxonomy. Here I have tried to bring them together, to provide a solid reference point for the many species and taxon pages where we are having to say that the taxonomy is about to go up in smoke because of molecular evidence. The dog/wolf example may seem inappropriately long, but I could not think of any way of presenting the technique other than by example. It would be good to add a second example illustrating a taxonomic revision that has been driven by molecular data.

The title molecular systematics might be questioned, because it is not widely used. But I can't find any other title that is more widely used, and the alternatives of molecular genetics and molecular evolution have to be rejected for the reason stated in the article - they cover a wider range of activities. Computational systematics is used, but less often, and in my view it is less satisfactory as a name - the fact that the technique is computationally heavy is an accidental rather than a fundamental characteristic.

Note that I am not a molecular systematist, nor even a systematist at all, and I have put together this page from reading I have had to do in order to understand a technique that turns up increasingly often in papers I have to read for other purposes. Real experts in the field are warmly invited to correct my errors and omissions - as if, within the Wikipedia world, they would wait for the invitation!

seglea 01:07, 25 Feb 2004 (UTC)


 * I just popped in to say what an excellent effort you have made to explain a very complex and often confusing topic. There is more to do (as there always is), but this fills a much needed hole. Good work! Tannin 07:33, 25 Feb 2004 (UTC)


 * er... thanks... I think... legend has it that my old professor once wrote a book review which stated simply that "This book fills a much needed gap", and on reflection the author didn't take it as a compliment... seglea 08:26, 25 Feb 2004 (UTC)


 * On much needed gaps : http://www.econ.uiuc.edu/~roger/gaps.html and http://www.ams.org/notices/199703/comm-mr.pdf
 * Though I am sure Tony meant it as a compliment! Pete/Pcb21 (talk) 09:54, 25 Feb 2004 (UTC)

To assert that "molecular systematics" is a cladistic approach is rather misleading, particularly if, like most biologists one assumes that cladistics=parsimony. While parsimony remains popular among systematicists in museums making classifications on the basis of morphological data, I would bet that the majority of molecular systematicists use statistical approaches. user:jhbadger


 * I've got "The History and Geography of Human Genes" sitting here in my lap and I can't help but sense weasel-wording surrounding their critique of maximum parsimony -- which is quite statistical when applied to nucleotide position characters. Aside from being riddled with questionable polemics such as referring to Ockham's Razor as "medieval theology" and frequently resorting to existential qualified statements about quantiative problems, like "The probability that the evolutionary path in gene-frequency space, will be the same as minimum path is practically zero", the critique relies on the fact that there may be multiple mutations at the same position leading to a loss of information, ie: you can't apply the "medieval" Ockham's Razor if the data is missing.  This is nonsense.  Missing data is inferred via expectation maximization which allows you to infer multiple single point mutations if it results in greater parsimony.  Indeed this is precisely why Ockham's Razor works: You can better infer expected results from simple statistical models than from complex models.  I really don't understand why people are so caught up in trying to set maximum parsimony up as a "medieval" paper tiger.  It's powerful and valid but you can't admit its validity if you can't allow it the benefit of statistical inference just as you can't admit any theory's validity if you can't allow it to deviate somewhat from real-world measurements that are certain to contain noise.  For a good discussion of the paper tiger see Sober's recent article "The Contest Between Parsimony and Likelihood" Jim Bowery

I think this is a great article - but poorly titled and in the wrong place. Most of the content (especialy the good sumary of the dog phylogey) ought to be merged with molecular phylogeny and placed under that title. Systematics is about classification, most scientists choose phylogeny as their criterion for classification, and where the primary criterion is a molecular phylogeny then classification becomes "molecular systematics" --DJO 11:04, 16 Apr 2005 (UTC)

I agree in full. Sometimes one is in love with new terms because one fancies the idea of progress. But there is a reason why terminology is strict and "unforgiving". What is explained here is molecular phylogeny, and molecular systematics might be an emerging sub-study which is performed by perhaps a handful of groups worldwide, but is original research. At least that`s my take.Slicky 14:47, 27 September 2006 (UTC)


 * I agree with the comments by DJO (11:04, 16 Apr 2005) and Slicky (14:47, 27 September 2006). Molecular systematics explicitly says the discipline is cladistic, and cladistics is a phylogeny-based classification system. Since "phylogenetics" tells informed readers up-front that it's about evolutionary family trees, its a more informative term than "systematics". Since "phylogenetics" is the study of "phylogeny" (the processes by which taxa evolve), I suggest the articles Molecular systematics and Molecular phylogeny should be merged under the title "Molecular phylogenetics". Philcha 17:35, 18 April 2007 (UTC)

Any chance some references could be provided for some of the statements made? Or even just a list of books/sites that were of use in researching the article?

clarification of terms needed?
The followinig comment does not bear on this article alone.

I don't know about others but I have a certain amount of confusion regarding the following terms:

Systematics, Taxonomy, Phylogeny, Phenetics, Cladistics, Scientific classification, Statistical classification

Do we need to do some work to make sure these concepts are all clear and reflect the understanding of most scientists?? I think we do. Jeeb 1 July 2005 04:02 (UTC)

Someone familiar with Vilà et al. (1997) should clarify result 4 under the dog phylogeny. Which clades are being referred to by "... the wolf haplotypes in these clades ..."? G Colyer 18:20, 14 June 2006 (UTC)

Molecular systematics: modern phenetics?
I would like to reiterate the point that jhbadger made on that "molecular systematics" is not particularly cladistic in principle, but with its focus on statistical methods should rather been seen as a modern form for phenetics. Fedor 07:41, 21 June 2006 (UTC)

I believe that phylogeny, taxonomy et al., are all part of systematics. Molecular phylogeny is a product of molecular systematics and it is plausible and certainly welcome that the two topics be merged into one. Thanks for such a great idea and the discussions are very informative