Time for a new tree, finally. Some groups have been fixed and the diagram has moved from Powerpoint to a real vector art program (Illustrator), so hopefully it looks a bit nicer now and has slightly fewer glaring errors. Have yet to fix all issues, the biggest (and hardest) being the proportions taken up by the various groups -- the tree appears dominated by Excavates for some reason. Due to lack of convenient taxa for the heteroloboseans and euglenids, I expanded them to the genus level in some cases to attempt to capture some of the diversity better, but that screwed things up for the rest of the tree. Since fixing that would require some hardcore structural changes to the whole tree, I'll do that later, in the next edition (which will not take over a year to come out this time). Given some conferences coming up this summer, and that people have asked, I'll release what I have done now as v1.3a. Enjoy! (And do complain if you spot anything awry...)
Previous versions and discussions, along with trees by other people, can be found here.
References
A shit ton (see image). But ResearchBlogging.org doesn't allow indexing 'shit ton', so I'm gonna be pathetically lazy and just cite this:
Keeling, P., Burger, G., Durnford, D., Lang, B., Lee, R., Pearlman, R., Roger, A., & Gray, M. (2005). The tree of eukaryotes Trends in Ecology & Evolution, 20 (12), 670-676 DOI: 10.1016/j.tree.2005.09.005
I've been doing my best to find errors, but (curses!) I can't spot anything significant. I shall be reduced to snivelling nit-picking...
ReplyDeleteThere should possibly be a '(p)' alongside Porifera, though to be honest the amount of actual work that's been done in this regard recently is pretty shameful. Morphological studies have assumed for a long time that sponges are paraphyletic; molecular studies either use demosponges only with no calcareans or hexactinellids, or (in the case of the one study I've seen including a calcarean) don't really test whether sponges are monophyletic or not (just assuming a sponge outgroup). Also, you're missing the Ctenophora.
Also, I'm really unhappy with whoever dubbed a certain group of undescribed hacrobians 'biliphytes'. TC-S had already introduced 'Biliphyta' a number of years previously as a name for a supposed grouping of Glaucophyta and Rhodophyta, and still continues to use it as such even while admitting its paraphyly. Even if you don't recognise the group directly, it's just confusing to try and use the same name for a completely different organism.
I do recall there being a debate about Porifera, but at the time (a year and a half ago) it seemed like the literature leaned towards monophyly. Personally, I'd prefer paraphyly because there's some potentially crazy rumours where sponge multicellularity is somehow fundamentally different from 'eumetazoan' multicel., and a poriferan paraphyly would conveniently put that to rest ;-) Maybe I'll add a (p?) for now. RE ctenophora, I'm not trying to be too comprehensive, particularly in the plants and animals, since there's already too many taxa there. I'd have to include placozoa as well if I include ctenos; I understand they're evolutionarily interesting, but are they a 'big enough' group? Maybe they are...
ReplyDeleteI did not know about Tom's 'Biliphytes', thank you! Maybe 'Picobiliphytes' (original term AFAIK), as awkward and unnecessarily long as it is, would be better. My boss uses 'biliphytes' so I kinda inherited that without thinking. There are growing rumours that they may not be '-phytes' after all, so hopefully that would lead to a whole new (sensible) name. Let's hide them from Tom... ;p
Hey! long time waiting for this! I'm glad you finally published it and looking forward for the next updated version after ISOP :P
ReplyDeleteShouldn't you place the polyphyletic-polytomic Hacrobia in between Archaeplastida and SAR? and in that way it would be easier to change those branches later :P
I really have some doubts regarding some of the polytomies you have there... I thought they were more or less resolved.
I'd have to include placozoa as well if I include ctenos; I understand they're evolutionarily interesting, but are they a 'big enough' group?
ReplyDeleteI think you could get away with including ctenophores but omitting Trichoplax adhaerens, seeing as the latter is only a single, difficult-to-place, described species (assuming, as seems likely, that the only-once-seen Treptoplax reptans is a misidentified Trichoplax). Ctenophores, on the other hand, are reasonably speciose and very abundant. In jellyfish season, you can find ctenophores around the place long after the other jellies have disappeared for the year.
Placement of putative roots plus showing deeper level splits in the tree as resolved can make things a little tricky looking. Currently, depending on how you look at the tree, you might view Excavata as being sister to the Opistokonts, depending on where you are thinking of the root being.
ReplyDeleteI'd also say it's still really really unresolved (and controversial) to definitively say there is a Red Algal endosymbiosis at the base of the SARH clade. There's some evidence for it being a possibility, but it is by no means definitive.
Excavates as sister to opisthos as opposed to Amoebozoa? I thought Unikonts were a pretty damn solid group, if not a fundamental split (ie unikonts-bikonts). I put multiple roots there, was going to add more sometime; thus far, the unikont-bikont root is the least extreme (Tom's "Eozoa" thing is just WTF on many levels at the point; who knows where the data will point though, Archezoa made perfect sense in the 90's); I agree with you RE chromalv hypothesis and single red algal symbiosis; I suspect there were probably two events, although Stramenos and Alveolates probably share a common one. Waiting until some stuff gets published though to update that part ;-)
ReplyDeleteThe tree is undeniably heavily influenced by a certain lab, and my own opinions are heavily biased, so it'll definitely be 'tainted' - creating a purely objective tree would be a nightmare and quite uninformative as there are more opinions than protistologists on this matter...
Which leads to Sergio's question about polytomies: the more you learn, the more of those you see ;-) One should assume polytomies by default rather than the other way around, one can't really doubt them, per se.
I completely agree that we should take polytomies as the default assumption, of course. I see you tried to be a bit conservative there, and also that you could be biased and thus just showing some protistologists' viewpoints.
ReplyDeleteIt is curious that more superficial clades tend to be more polytomized than deeper ones, anyhow I understand it would be quite uninformative if you don't stick to a particular school.
Overall, I really like your tree and the fact that is constantly under revision.
Sorry should have worded that better, or more explicit. If you did the putative root on the branch leading to Archeaplastida, which there is one drawn on there, you would get Excavata as being more closely related to Opisto's + Amoebezoa. It's just a function of drawing resolved branches at the deeper nodes but having more than one potential root indicated, with the tree being actually drawn unrooted. Or rather the tree could be easily misinterpreted at first glance as showing that (to the uninitated). Most people can't decipher trees properly but that's hard to adjust for.
ReplyDeleteRE: Chromalveolates: I agree that it is pretty much impossible to make a purely objective tree. We've always leaned though towards making it known when something we agree with isn't really firm. So if I was drawing the tree I'd have a question mark beside the red algal endosymbiosis to denote it's uncertainty. Simple things that don't really have to make the tree more complicated.
Congratulations on the SciAm blogging - when I started reading down Bora's list I immediately thought that this was where you would belong.
ReplyDeleteI'm leaving a comment here because the Scientific American post won't let me comment unless I register. Even though I can comment without registering at Tet Zoo.
ReplyDelete...fungus (defined by…let’s not go there today, but roughly speaking, chitinous walls and presence of a dikaryotic life cycle stage, which I believe is in at least most of them).
Tough, you can go there ;-P . The dikaryotic stage is found only in one subclade of the fungi called, funnily enough, the Dikarya, that includes the ascomycetes and basidiomycetes. The 'chytridiomycetes' and 'zygomycetes' don't have a dikaryon. I believe the chitinous walls are still universal for fungi, though (except perhaps in a few cases where they've been secondarily lost).