A fly by any other name
Sarah Kemmitt
Friday, 06 February 2009 15:58 UTC
A recent article in Nature (20 January 2009, doi:10.1038/457368a) details a debate about whether to reclassify Drosophila. Some argue that it would have repercussions with respect to literature discovery because of the old name not surfacing when searching under the new name. Others argue that there is sufficient linking and sophistication in search engines for this not to be an issue. Whatever the case, it is already an issue as most organisms seem to have multiple synonyms anyway. Does moving over to a system of GUIDs (e.g. LSIDs) overcome this problem or not? Presumably it is still important to know where these organisms come on the ‘tree of life’, whatever naming system is used. Or is the whole tree of life concept redundant as suggested in the recent New Scientist article?
-
Replies
-
Lots of issues here. I haven’t read the Nature piece (it’s not free, so no access from home), but the challenge of finding information linked to multiple names is often referred to as “taxonomic intelligence”. GUIDs will only help there are databases that in the GUIDs (e.g., that can tell you that the name identified by one GUID is a synonym of the name identified by another GUID).
The New Scientist piece on the tree of life is old news, and not terribly sophisticated. Mathematically one can show that even in the face of reticulation and horizontal gene transfer it can still make sense to talk about a tree (see e.g. Mike Steel’s chapter in Reconstructing the Tree of Life: Taxonomy and Systematics of Species Rich Taxa (a PDF is available from Mike’s web site).
That different genes yield different trees is not news, but recombination is not the only explanation (gene duplication and loss, and coalescent process are also involved, e.g. doi:10.1016/S0169-5347(98)01438-4).
I’d query the New Scientist’s statement that
The vast majority of eukaryote species are unicellular – amoebas, algae and the rest of what used to be known as “protists”
The vast majority of (described) species are insects, although these may well be one day overhauled by nematodes and microfungi. The unicellular eukaryotes represent a much greater span of genetic diversity than, say, animals, but not AFAIK a greater range of taxa.
-
There is a simple philosophical question here. In order to change the name of something it has to have an identity separate from the name itself. i.e. there is some delimitation of Drosophila somewhere that is separate from all the observations that have been tagged with the word ‘Drosophila’. It is saying ‘Drosphila’ is an attribute of this immutable thing – the taxon. What you seem to imply is that if we make up a new name – Drosophila2 – all the prior observations that were tagged as Drosophila1 some how apply to it because the taxon is an imutable thing we all know about and all we have to do is get our search engines to find things tagged with both names. All we have done is change the name. Think of a dictionary with a single definition of a word and we simply come up with a new string of letters to represent that word.
This is the most fundamental error people make about taxonomy. Names are not the same thing as taxa. Names have no volume. We can’t simply do query expansion on the names and some how the results will all work out wonderfully.
No one would think we can translate from one human language to another simply by doing a word to word mapping in the dictionaries of the two languages. The same is true of taxonomic classifications. You can’t just map from the names used in one to the names used in another because they have the same type specimens. The meanings (i.e. the taxon circumscriptions) may be very different in the different classifications.
The issues of GUIDs come in when we tag the meanings (the definitions of the words in the dictionaries) rather than the names. We can then make statements about the relationships between taxa rather than between names.
If taxonomy is in a crisis it is for the single reason that there is no formal way of declaring a preferred circumscription of a taxon. Where, for example, is their a formal definition of what Drosophila actually is. How different does a fly have to be to not be Drosophila. How do I decide whether observation A was made on the same taxon as observation B and can therefore combined in my analysis if I don’t know how the scientists who made the observations were using the same circumscriptions of the organisms. They may have been using the same words but with different meanings.
-
In the specific case of Drosophila melanogaster mentioned in the Nature piece (http://dx.doi.org/10.1038/457368a), it really is simply a matter of names. Recent phylogenetic research suggests that we should be calling this fly Sophophora melanogaster, which could be massively disruptive (imagine changing our name from Homo sapiens). There is a proposal before the International Commission on Zoological Nomenclature to conserve the name Drosophila melanogaster (see Splitting the genus Drosophila). Hence, this is simply about what is the appropriate name. For example, the URL for the Wikipedia page for Drosophila melanogaster http://en.wikipedia.org/wiki/Drosophila_melanogaster could be changed to http://en.wikipedia.org/wiki/Sophophora_melanogaster, all references to Drosophila melanogaster changed to Sophophora melanogaster, and we’d be done,
Roger makes a broader point, which I take to be that how we know that all the observations of “Drosophila melanogaster” reported in the literature (and, for example, listed on the Wikipedia page) actually refer to the same thing? Different people may have different notions of the scope of that taxon. For example, in 1919 Sturtevant described Drosophila simulans (http://dx.doi.org/10.1155/1919/97402), a sibling species of Drosophila melanogaster. This species was discovered among stocks regarded as Drosophila melanogaster, hence observations on Drosophila melanogaster prior to 1919 might well have been referring to Drosophila simulans.
Regarding the question of “Where, for example, is there a formal definition of what Drosophila actually is”, there are some ways we could provide one. For example, advocates of DNA barcoding might argue that we could define taxa as clusters of DNA sequences of a given degree of similarity, and the probability of membership in a given cluster could be computed fairly easily. Advocates of the Phylocode could argue that we could provide a definition of Drosophila such that, given a phylogenetic tree, we could unambiguously decide whether a taxon belonged in Drosophila.
-
Hi Rod,
I should shut up now as I haven’t actually read the paper (no access) and Drosophila is probably a crazy taxon to use as an example but I’ll just follow up on a couple of your points.
1) There is a basic principle that part of the circumscription of a taxon is inherited from the parent taxon. Sophophora has to have a different circumscription from Drosophila because it excludes the type of Drosophila. If S. melanogaster inherits the characteristics of Sophophora and D. melanogaster inherits the characteristics of Drosophila they must be different. This sounds incredibly pedantic when we are talking about a model organism like D. melanogaster but not for the millions of other organisms.
You could deny that taxa are circumscribed by the characteristics of their parent taxa and it would certainly make life easier from an informatics point of view but I believe this is a radical departure from how taxonomists work. The majority of fauna, flora or monographs are constructed by defining higher taxa and then differentiating between their members.
2) The point that I make concerning formal definitions isn’t that we couldn’t do it if we chose to but that we don’t do it. It isn’t even seen as important because most people don’t see that the use of names can be ambiguous.
Thanks for expanding my point with D. simulans. It illustrates it well.
-
Roger,
I think we may be at cross purposes (or maybe I’m being dense). The issue regarding Drosophila melanogaster is, I think, purely one of nomenclature. If we break up the genus Drosophila, then as Drosophila melanogaster is the type species of Sophophora, it becomes Sophophora melanogaster. It’s not an issue of what Drosophila melanogaster is, rather it’s what to call it. The proposal before the ICZN is to make Drosophila melanogaster the type of the genus Drosophila, keeping happy the people who study the fly itself.
I think I see your point about higher taxa, but I think the extent to which a higher taxon circumscribes a lower one is an empirical question. In many cases there isn’t a character shared by every single lower taxon (e.g., snakes are tetrapods [“four legs”] without legs).
Perhaps the bigger issue is how closely we should attempt to mirror the way we think taxonomists work. I’m not convinced this is the best way forward.
-
Hi Rod,
I have finally read the article and I think it illustrates the one key thing we agree on – and the title of the forum. Here are a bunch of people spending a lot of time arguing about what to call a taxon purely because of the binomial naming system. Surely their time would be better spent doing biology! If biologists are wasting time arguing about what to call something then taxonomy is in crisis. Naming should be a no brainer service. Our current nomenclatural systems are not fit for purpose in the “information age”.
(I still maintain that I can be pedantic enough to claim not to know what the ‘thing’ they are talking about actually is – there is no formal definition so I guess it is just whatever I get back from Google when I enter Drosophila melanogaster)
Roger
-
Actually, part of the issue with the Drosophila case in particular is that it isn’t just the philosophical and nomenclatural questions at stake, but simply the basic science behind it. While the genus Drosophila is unwieldy and clearly paraphyletic with respect to some of the described genera, it’s not clear how it should be divided up. Some of the new proposed genera are not definitely monophyletic themselves and don’t have diagnostic characters. If it were a more clear-cut case, I think there would be less of an uproar. It gives the appearance of not only pushing a major nomenclatural upheaval onto an important group, but doing so without the solid evidence that should be required for something like this. As a result, that makes it much more difficult for anyone to come along later with good data and arrange things properly.
As an aside, I think it’s a bit ironic that in the Nature article Patrick O’Grady cites Aedes/Stegomyia as an example of how name changes can cause confusion, while one of the BZN comments cited it to say that changing to Sophophora melanogaster wouldn’t be such a big deal.
-
Actually, paraphyly of Drosophila is not a problem but a wonderful opportunity to examine descent with modification of taxa. Clearly the paraphyletic group is the ancestor of the autophyletic taxa.
It is only an importunate homage on the part of phylogeneticists to their sister-group analytic method that requires enforcement of strict phylogenetic monophyly (holophyly) in modern classifications, splitting, excising, or reducing in rank taxa that should have unique evolutionary traits flagged at an appropriate level in classification. Because ancestor-descendant relationships are not recognized in phylogenetic classification, a major source of evolutionary information is gradually being deleted from the classifications that inform biodiversity and evolutionary study. This isolates systematics, which presently cannot or will not provide a general-purpose classification.
I’ve said this elsewhere as a cross posting to the listserver Taxacom, so far to no effect.
-
There are two issues. One is the taxonomic issue, and the second is online database issues.
Online database issues are relative easy, as a hit for a specific search term can point towards another term. That requires just a single table in a database, with search term to actual term. Google is using this at places.
The second issue is the naming of Drosophila melanogaster, and the usual misconceptions present there. The current type species of the genus Drosophila is only distantly related to Drosophila melanogaster. Any taxonomic revision that separates Drosophila melanogaster from Drosophila funebris, results in the renaming of Drosophila melanogaster. The question is whether or not that is desirable. After that issues is resolved, revisions of the genus can take place, without having to deal with the naming of this important model organism, and alternative proposals to deal with the genus can be proposed based on differences of opinion.
-
