…and this is a 
I use a very pretty radial tree of life diagram fairly often — the last time was in my talk on Friday — and every time I do, people ask where I got it. Here it is: it's from the David Hillis lab, with this description:
This file can be printed as a wall poster. Printing at least 54" wide is recommended. (If you would prefer a simplified version with common names, please see below.) Blueprint shops and other places with large format printers can print this file for you. You are welcome to use it for non-commercial educational purposes. Please cite the source as David M. Hillis, Derrick Zwickl, and Robin Gutell, University of Texas. About this Tree: This tree is from an analysis of small subunit rRNA sequences sampled from about 3,000 species from throughout the Tree of Life. The species were chosen based on their availability, but we attempted to include most of the major groups, sampled very roughly in proportion to the number of known species in each group (although many groups remain over- or under-represented). The number of species represented is approximately the square-root of the number of species thought to exist on Earth (i.e., three thousand out of an estimated nine million species), or about 0.18% of the 1.7 million species that have been formally described and named. This tree has been used in many museum displays and other educational exhibits, and its use for educational purposes is welcomed.
There's also a simplified version:
Both of those are available as scalable pdfs, so you can zoom in and out to get just the right view, which is very handy.
Life Science
Comments
Posted by: Zeno
|
June 20, 2010 1:53 AM
Nice!
Posted by: B166ER
|
June 20, 2010 2:05 AM
Such a rad diagram and it's educational too! I especially like the "you are here" part in REAAALLLYYY tiny print off to the left side. Great find Prof. Myers, I will definitely save this and use it in any educational ways possible!
No Gods, No Masters
Cameron
Posted by: Wowbagger, Man-Hating Man of Pharyngula
|
June 20, 2010 2:05 AM
Damn, I thought from my initial glance at the first picture that you were announcing your plans to build yourself a Death Star.
Posted by: AnneH
|
June 20, 2010 2:07 AM
I've clicked through to the site, and I recommend others do so. My favorite version is the one in a tree, although I think the tattoos are almost as awesome.
I'll pass this on to the biology teachers at my high school just before school begins in August.
Posted by: ArmandTanzarian
|
June 20, 2010 2:07 AM
Does anyone have a nice wallpaper version of this? Thanks a bunch.
BTW coincidentially I just finished reading Dawkins' The Greatest Show On Earth. That picture was in it, as is that wonderful Clare D' Alberto tattoo.
Posted by: mattchewd
|
June 20, 2010 2:13 AM
I had my dad (architect) print out a poster of this radial tree. It's really cool, but I wish that the Bacteria and Archaea weren't so underrepresented.
Posted by: R. Schauer
|
June 20, 2010 2:31 AM
A few months back, I saw this image or a similar one on a T-shirt someplace. (...searching...)
Posted by: Your Mighty Overload - Un-True Scotsman
|
June 20, 2010 2:31 AM
Actually, I'm writing my new first year Biology classes at the moment, and I use both these diagrams in my slides dealing with taxonomy. Mainly so I can put it on screen, and say to the students, "okay, I'll give you 5 minutes to copy this down". =P
I use Sandra Baldauf's scheme (2008; Journal of Systematics and Evolution) for taxonomy of Eukaryotes. Baldauf's seems to be currently the most complete.
Posted by: Your Mighty Overload - Un-True Scotsman
|
June 20, 2010 2:39 AM
Mattchewd
Yeah, you're right about the bacteria and archaea being under-represented, but even within the Eukaryota, the vast majority of species are unicellular (i.e. protists - which is actually not a taxonomic group anymore). Ahh, it's all changed since I did my degree - and it's less than 13 years since I started my undergrad!
Posted by: humanizzm
|
June 20, 2010 3:28 AM
Thanks alot for the link! This is great, I'll get my poster printed on monday.
Posted by: toomanytribbles
|
June 20, 2010 4:00 AM
i've loved the hillis plot since i first saw it on your blog, i believe. it's so... egalitarian!
Posted by: jdmuys
|
June 20, 2010 4:24 AM
Wow, this is nice. Yet it leaves a number of questions open that I'd like explained.
As a layman in biology:
does the order around the circle have any significance?
does the circular representation have a meaning or is this "only" a nice visual gimmick?
As a computer programmer:
how was it done?
on which machine?
with which language?
starting from which data set?
is the data set available somewhere for download to play with it?
Posted by: MetzO'Magic
|
June 20, 2010 5:02 AM
And... the text is the diagramme is searchable. We share a LCA with the European rabbit and the noble brown rat. Nice.
Posted by: skeptical scientist
|
June 20, 2010 5:09 AM
This visual depiction of a representative sample of all life on Earth really drives home what a small percentage of all life is comprised by the animals we see on nature shows (mostly tetrapods, a few fish). We're only four names away from alligators!
jdmuys, I'll try to answer some of your questions:
does the order around the circle have any significance?
Yes and no. It's a representation of a tree, so closeness has some meaning—for example, all of the mammals are grouped together, because we all have a common ancestor which is more recent than the last common ancestor of mammals and reptiles, and the rodents are grouped together within the mammals. However, there are many degrees of freedom within this representation—for example, we could have switched places with the rodents without violating any of the geometry, but we couldn't have switched places with e. coli.
Basically, you can linearly order the leaves of a tree in such a way that all descendants of a single node are together in a single interval. But there are lots of ways to do this, and I'm not sure how the author of this diagram picked this particular one.
does the circular representation have a meaning or is this "only" a nice visual gimmick?
The fact that it is a circle doesn't really have any meaning. As I said above, it's essentially a linear order (and even that is somewhat artificial), so the fact that the two opposite ends of the linear order are next to each other is fairly meaningless. (But so is the fact that homo sapiens happens to be in the middle of the animals segment, while the fruit fly is at the end next to fungus. After all, we're just as as close as the fruit flies to fungus.) The primary reason for it being a circle, apart from the aesthetic benefit, is compactness. Trees tend to grow really fast, exponentially fast in fact, as I'm sure you know as a programmer. So each generation wants to be a lot wider than the previous one, which is why trees spread as they grow, and the best way to represent this graphically in two dimensions is with a circular layout with the root at the center.
Posted by: Escherichia coli
|
June 20, 2010 5:26 AM
jdmuys @ #12,
The order around the circumference doesn't really matter. What you need to do is to follow the branches (the lines that radiate from the centre and fork out). You can swap the position of each fork and it still means the same thing.
Example:
| |
| |
/ \ is the same as / \
/ \ / \
A B B A
Yeah it's sort of a gimmick. You can display it in a more traditional "tree" fashion too. Basically a personal preference, but reading it is the same either way.
According to the linked page, the authors used the ribosomal RNA (sort of like a less stable cousin of DNA, and generally more functional) of each species' ribosomal small subunit. How they did it is probably by looking at the differences of the RNA sequences between the species, then grouped the least dissimilar ones together.
Eg if you have 3 sequences,
1) AAGGCUCCA
2) AAGACUCCA
3) UUGACUCCA
1 and 2 are most alike, so they are most closely related. The branching is therefore most recent (nearest to the circumference).
Not too sure what database they use, but you can try experimenting with other kinds of data (eg protein sequence, genomic DNA) and generally get the same tree. There are various databases available, but a good place to start would be the NCBI database: http://www.ncbi.nlm.nih.gov/Entrez/
For an easy program to fool around with, you can try this:
http://www.megasoftware.net/
Posted by: Escherichia coli
|
June 20, 2010 5:29 AM
Argh, sorry for the failed drawing. The preview displayed it fine :( Hope it's still comprehensible.
Posted by: johnred23
|
June 20, 2010 5:58 AM
Ooh, I think I just found my next computer wallpaper!
Posted by: Vilding
|
June 20, 2010 6:16 AM
Us, house mice, Norway rats and rabbits. Couldn't they have thought of more interesting representatives for the mammal group? Well, Norway rats are cool, I guess.
And perhaps it is too early in the morning, but where are the cephalopods?
Posted by: yeglit
|
June 20, 2010 6:27 AM
Gah, the diversity of the groups is COMPLETELY BACKWARDS! To say it's skewed is a major understatement. Sure, it's beautiful, but considering that protists and bacteria make up more than 99% of diversity, not the tiny sliver they're given on that diagram, it makes me cringe. A lot. (within Eukarya alone, protists make up over 99% of the diversity...)
For better (more accurately sampled) trees, see Wu et al 2009 Nature (Eisen's GEBA tree*) and Keeling et al 2005 Trends Ecol Evol (eukaryotes); as well as the Cicarelli tree (though the eukaryotes are screwed up there). Baldauf's tree is ok, but we have some disagreements with them so nyeh =P
*high res version ranks high on the sexiness scale, btw
Incidentally, I've been working on a tree of eukaryotes myself; it has a few errors here and there but doesn't make the experts cringe in agony =D Didn't include bacteria because a) their phylogeny is still mostly a total mess; b) then there'd be no space for eukaryotes =P
I have citations by the relevant nodes in the tree, which I think may sometimes be more useful than the tree itself:
A Tree of Eukaryotes v1.2
The plan is to someday spice it up with pictures. When I have time...
[/shameless advertising]
@jdmuys Order does NOT matter, it's the branchings themselves that do; ie (a,(b,c)) is the same as ((b,c),a) but not the same as ((a,b),c) if commas indicated branches and brackets indicated clades ((a,(b,c)) means b and c are closer to each other than either is to a)
Phylogenies are not very text-friendly...
If you'd like to play around with trees yourself, go to www.mesquiteproject.org by Wayne & David Maddison. Fun software! More userfriendly than Phylip and others, though also more memory-taxing, being Java based.
What is cool that with a bit of training and understanding, you can basically take freely-available sequences from GenBank (obviously while citing them rigorously), run them through free online servers and build your own trees!
Posted by: scooterKPFT
|
June 20, 2010 6:28 AM
I bought one of those spiral versions of this diagram from the guy at the Burbank AAI last year, it's hanging on the wall in our 'family room', i.e. where all the computers are.
Posted by: yeglit
|
June 20, 2010 6:31 AM
Ack, link didn't work for some reason:
http://skepticwonder.fieldofscience.com/2010/03/tree-of-eukaryotes-v12.html
A Tree of Eukaryotes v1.2
Posted by: 'Tis Himself, Quel Dommage
|
June 20, 2010 6:46 AM
Biologists have all the neat stuff like circular tree diagrams and stuff like that. Us social scientists don't get any good stuff.
*Snivel*
Posted by: Alice Bluegown
|
June 20, 2010 7:18 AM
This is, of course, utterly awesome - having just completed 'The Ancestor's Tale' I am getting used to the idea of radial trees, and one on this scale is just mind-blowing. About to follow Armand Tanzarian's lead and embark upon 'The Greatest Show on Earth' - can't get enough Dawkins!
Posted by: dpattersonmonroe
|
June 20, 2010 7:28 AM
Very nice! Definitely going on the wall in our homeschool - thanks!
Posted by: McCthulhu is taking ∞ to eat all the pi
|
June 20, 2010 7:44 AM
Here's to radial wheels replacing crosses, stars and crescent moons everywhere. The view back to the center of the wheel from our tiny sliver of pie is dizzyingly awesome when you think of the time and events that had to take place to make it happen. Our cousin, the cyanobacteria had no comment, however.
Posted by: morgan in austin
|
June 20, 2010 7:52 AM
From your blog to the classroom wall.
I also like this tree-structured diagram, especially since it shows the "kingdoms".
http://www.discoverlife.org/mp/20m?tree=Life&res=1200&flags=all:&b=WHF_LIFE
Posted by: morgan in austin
|
June 20, 2010 8:05 AM
Yeglit wrote
Links
Eisen/GBA/Nature
http://www.nature.com/nature/journal/v462/n7276/full/nature08656.html
Keeling/Trends Ecol Evol/eukaryotes
http://sbli.ls.manchester.ac.uk/fungi/21st_Century_Guidebook_to_Fungi/REPRINT_collection/Keeling_etal_tree_of_eukaryotes2005.pdf
Posted by: https://me.yahoo.com/a/O6jE0BwUkIDyMwTXUQbv1j_oNGhXtSBB#5c01e
|
June 20, 2010 8:06 AM
I was looking at one of these diagrams at Wikipedia the other day. They indicate that all life sequenced so far shares common ancestry, correct?
Is the current consensus that there was only one 'seed' life form that all life on Earth evolved from? If so, do we think the initial formation of life is exceedingly unusual, or did other 'seed' life forms perhaps appear but not survive?
I realize this is speculative, I just wondered what people who study this sort of thing feel is likely.
Posted by: jdmuys
|
June 20, 2010 8:09 AM
Thanks all for the information. Today is a good day: I learned many things.
Posted by: daveau
|
June 20, 2010 8:10 AM
I love the "You are here."
I think it's wrong. I thought humans & cephalopods were more closely related than that.
Posted by: Andrew Hall
|
June 20, 2010 8:31 AM
This tree is missing the Nephlim, the race of giants spawned by angels and man.
I thought I'd bring it to your attention.
http://laughinginpurgatory.blogspot.com/2010/06/im-wondering-what-are-peoples-thoughts.html
Posted by: davelnewton
|
June 20, 2010 9:04 AM
Ah, I was wondering what to put on my elbow instead of a stupid spider web--now I know.
Posted by: McCthulhu is taking ∞ to eat all the pi
|
June 20, 2010 9:07 AM
If you get into a 'debate' with a cdesign proponentsist (aka the BOTHER-in-law) about these diagrams and what they mean and you happen to blurt out "Just because you failed your science courses or never bothered to take them doesn't mean you should take it out on yourself with religion!" they will call you rude. Interesting bit of data that.
Posted by: Sven DiMilo
|
June 20, 2010 9:08 AM
The circular shape is for space-use efficiency; it's just a regular bifurcating tree rolled up. The ONLY information contained here is the order of branching, and it's very important to emphasize that all of the branching patterns shown are hypotheses, not Truth. Some are better supported than others, and many are going to end up wrong as more data are applied to the questions.
Nah. Systematics these days is almost all done by cladistic algorithms--species are grouped not by overall similarity (or lack of dissimilarity) but rather by reliance on especially informative shared derived traits. The general idea (and I am not a systematist) is to look for specific evolutionary changes (trait differences between groups) and group organisms that share the inferred derived (most recently evolved) traits. It gets very complicated very quickly from there.
and yeah, I'm not certain, but there don't seem to be any cephalopods, despite a ridicuous number of pulmonate snails...
Posted by: scooterKPFT
|
June 20, 2010 9:58 AM
Is my brain farting or par the second diagram, did all life begin with spirochaetes and chlamydia? What an unpleasant thought, two of our least favorite critters.
Posted by: Sven DiMilo
|
June 20, 2010 10:11 AM
No, modern spirochaetes and chlamydias are at the tips of the earliest illustrated branches.
Another way to say it: the ribosomal RNA of modern spirochaetes and chlamydias has (apparently) accumulated the fewest changes (from the ur-ribosome) of all the modern organisms pictured (all of which occupy the tips of equally long (in time) total branch-length).
Posted by: Antiochus Epiphanes
|
June 20, 2010 10:16 AM
What Sven said, but I would add that this tree is based (AFAIKT) on a single gene, and probably required methodology even more complex than cladistics*. The Science paper that the Hillis site refers to is a news article, not a peer-reviewed doohickey, and in my current state of flux, is hidden from me by a paywall. Other things: The origin of the eukaryotes was certainly not a branching event--this is a gene tree, and not an organismal tree, so there is another layer of inference there.
And also, I am grumpy and need more coffee and shouldn't be thinking about trees until I have had some.
*In short: This tree is a fun tree, but by no means a well-tested hypothesis of evolutionary history.
Posted by: bhoytony
|
June 20, 2010 10:18 AM
Why would you expect to see cephalopods? Everyone knows our tentacled overlords come from Beyond Space And Time.
Posted by: Sven DiMilo
|
June 20, 2010 10:20 AM
? You're talking about an endosymbiotic origin of the nucleus? (I'm not up on my deep branches.)
Posted by: Antiochus Epiphanes
|
June 20, 2010 10:24 AM
Yeah. I'm considering the eukaryote to comprise a protobacterium (mitochondrion) plus (possibly) a highly bombarded Archaean (nuclear precursor).
To say nothing of the later cyanobacterial buddy (chloroplast).
Posted by: John Harshman
|
June 20, 2010 10:32 AM
No, it isn't that either. This tree has no information on numbers of changes. What it means is that the earliest branching of life (at least that's represented here, and as claimed by this analysis) separates bacteria from all other taxa, while the earliest branch within bacteria separates spirochetes from all other taxa, and the second earliest separates chlamydia from all others. It's just that the tree happened to be printed with spirochetes farthest to one side and birds farthest to the other side; as others have explained, that's an arbitrary situation, and like a Calder mobile, any part of the tree can be rotated about any branch junction without affecting its information in any way.
Posted by: Not Guilty
|
June 20, 2010 10:32 AM
New desktop background in 3...2...1...
Posted by: raven
|
June 20, 2010 10:32 AM
What the data says. The genetic code is universal.
No one knows right now. It was 3.7 billion years ago at least. Very little of the earth's crust from that era has even survived.
1. One hypothesis states that earliest life was collaborative and not overly cellularized. The clade that survived could represent a coalescence of many types that gained a competetive edge and took over.
2. Someone once calculated that life evolved on earth between 1 and 3 times. It is possible that there were other independent abiogenesis events and ours ate them. This isn't much better than a wild guess without any data.
Posted by: Anubis Bloodsin the third
|
June 20, 2010 10:41 AM
As noted earlier in the thread, we, as in Homo sapiens, are approx three branches from Rattus norvegicus (Brown Rat) and mus musculus (House Mouse) in this tree!
How comes that does not overly shock me ?
Homo Sapiens next to vermin, sounds reasonable methinks!
Posted by: Algernon, elle sans chapeau
|
June 20, 2010 10:41 AM
Wow, this really is pretty. I love information organization as art.
Wow... that was nerdy.
Posted by: Antiochus Epiphanes
|
June 20, 2010 10:45 AM
Anubis: Mammals are necessarily too inclusive to be well sampled?
Posted by: https://www.google.com/accounts/o8/id?id=AItOawmqD_mcUIrSfOTlK3iGVsnEDcZmI43srbI
|
June 20, 2010 10:50 AM
Which ones are the baramins?
Posted by: Glen Davidson
|
June 20, 2010 10:53 AM
Notice that someone had to design it.
As the IDiots would say, therefore God -- or, the "Designer" -- depending on who they were lying to at the time.
Glen Davidson
Posted by: Sven DiMilo
|
June 20, 2010 10:55 AM
Right--thanks for the correction.
Posted by: Antiochus Epiphanes
|
June 20, 2010 11:07 AM
I would point out also that there is a tremendous amount of horizontal gene transfer among the prokaryotes...probably no reasonable explanation of tree-like evolution there either.
Posted by: Anubis Bloodsin the third
|
June 20, 2010 11:09 AM
#46
Seems the irony would be just to ironic for the Messrs. Ham & Hovind's of this world.
Considering the diagram is reckoned to be approximately a square root of the species count so far...3000 odd samples against the 9 mill thought to be swanning about today...Just thought it a just and philosophically correct placement.
But admit to bias!
Posted by: Ibis3, féministe avec un titre française de fantaisie
|
June 20, 2010 11:11 AM
I wonder how big the circle would have to be to show every species?
Posted by: Tulse
|
June 20, 2010 11:18 AM
Now that is a great idea for a science museum display...
Posted by: AnthonyK
|
June 20, 2010 11:24 AM
There's an even better, even cooler, mammailian version of this on the cool informatics site.
And there's a rather nice TV snippet of someone wandering around on it, too.
But re the tree of life in this post - why are we placed next to rats and mice? Is this just a whimsical cladistic joke, or is there any significance to it?
Posted by: Sven DiMilo
|
June 20, 2010 11:33 AM
There are only 4 mammals in the analysis. You like rabbits better?
Posted by: Richard Wolford
|
June 20, 2010 11:37 AM
Is there a simplified version with pictures? That would make a great poster for my 4 year old, plus it can be used with creotards (they respond better to pictures, reading is hard).
Posted by: Tigger_the_Wing
|
June 20, 2010 11:41 AM
Ibis3 @ #52
Well, if we assume approx. 2mm of circumference for each word (just about legible), the above tree could be printed out at about 1.9 metres across.
With the 1.7 million named species, the circle would be over a kilometre across.
With the 9 million species it is conjectured actualy live on our planet, the circle would be about 5.7 kilometres across.
Posted by: AnthonyK
|
June 20, 2010 11:50 AM
Richard - try the link above. The 41/2 minute film accompanying it is very instructive, and the tree of life itself is full of familiar creatures (and kind of explained my question above).
Posted by: Sastra
|
June 20, 2010 11:56 AM
I have this diagram (or one like it) on a light blue t-shirt (and on the back there's a magnification of the animal section.) Can't remember where I bought it, but it's probably on-line.
Despite its flaws, it's a nice counter to the Great Chain of Being concept. The other day, in a shop window, I a t-shirt with the typical fish-mammal-ape-caveman-modern man walking figures, with the guy at the end turned around, complaining "Hey! Stop following me!" Since PZ hates that popular design (implies progression), I assume that last figure represented him.
I thought about buying it for him and giving it in Montreal for the AAI, but I suspect he has a whole closet filled with such t-shirts, given by people who see things and think hey, I bet he doesn't have this! Better to buy him a drink.
Posted by: steinman.bio
|
June 20, 2010 12:02 PM
This is VERY cool. I am definitely printing this out and posting it in my classroom!
Posted by: FrankO
|
June 20, 2010 12:27 PM
#43
Or to be pedantic: the genetic code is generally universal, but quite a few organisms show variations on the universal theme. When this happens there is good evidence that it resulted from a gradual evolutionary shift away from the 'universal' code, rather than originating from a separate first organism, as asked by #28.
Posted by: the__eye
|
June 20, 2010 12:34 PM
We have a giant version of this on the wall outside the first year bio labs at my uni. It fascinates me.
Posted by: https://me.yahoo.com/a/O.jullMj0I2VvJaxMMVeNKSfOPf73voLSxJAe9PdlOWwi8Y-#258ec
|
June 20, 2010 1:32 PM
I think that the circular design of "the tree of life" does indicate visually that life radiated from a "first life" better than a more "traditional" tree trunk and branches does. The traditional tree and branches to me implies an upward progression to the development of life when what actually seemed to happen was a radiation from a beginning. A more accurate representation might be better done in three dimensions which would not make it easier to make but smaller if it was to be include all living species living today, it might even be possible to indicate the times involved better. a physical model in that form would be very hard to make but a digital one might be some what easier. Just what my head does when it aint doing anything important just thinks up variations.
now for some more coffee and something yummy!
uncle frogy
Posted by: https://me.yahoo.com/a/K2PNji0at.txAjzTShOlxwLuFcVVFwbnng--#bd813
|
June 20, 2010 2:01 PM
The 'one germ' idea has bothered me for a long time. I'm thinking now, one line of pre-cellular replicators got an edge on the others, and ate 'um up.
Posted by: mdcaton
|
June 20, 2010 2:52 PM
Here's a question - and keep in mind I'm bioinformatically challenged or I probably wouldn't have to ask it. Is there a way to build a radial tree for all the transcribed genes in a single organism (assuming that many or most will be descended from other genes by duplication events)? What would be really interesting would be if such a tree could label how long ago or what ancestor organism it was when the split occurred (i.e. "oh look, the glutamate and glycine neurotransmitter receptors branched off from each other in a C. elegans-like ancestor"). Thanks in advance.
Posted by: Bryan Foster
|
June 20, 2010 3:20 PM
I like
Posted by: ronsullivan
|
June 20, 2010 3:38 PM
A three-dimensional version of one of these inside a transparent sphere might be an interesting piece of art.
Posted by: AnneH
|
June 20, 2010 3:54 PM
Radial images seem to be a theme on the blogs I visit today. Here's a pretty picture of color varieties in carrots-
http://www.boingboing.net/2010/06/20/carrot-rainbow.html
Posted by: McCthulhu is taking ∞ to eat all the pi
|
June 20, 2010 4:09 PM
@#36: I may just be experiencing a brain cloud here, but I seem to recall an article with the conjecture that larger life forms may have come from virus-like critters. Would bring a whole new dimension to the old 'we have met the enemy and it is us,' saying. Goes a long way in explaining our propensity to eat/poop/reproduce ourselves beyond the means of the local environment to sustain ourselves too.
Posted by: monad
|
June 20, 2010 4:54 PM
@9 Your Mighty Overload:
Right now there are way more recognized species of insects than there are of anything unicellular. Probably this is partly because we don't know protists as well, but I wonder if it isn't partly because they don't tend to divide into specialized groups recognizable as "species" to the same extent.
@50 Antiochus Epiphanes
There's a lot of horizontal gene transfer, but is there enough that there isn't a set of "core genes" reflecting tree-like descent? Some biologists have said yes, but others have said no. It seems to me that prokaryote phylogeny still relies mostly on rRNA, and since that gave such misleading results for eukaroytes, we'll have to wait for a better picture.
Posted by: excentricat
|
June 20, 2010 4:59 PM
For those who want a poster, but don't want to spend the money to have it printed at 54", Kinkos or other places can print at 36" for way less. It is readable at this if you have good eyesight. I also keep a magnifying glass near mine for harder to decipher words.
Posted by: Owlmirror
|
June 20, 2010 5:27 PM
You've received some answers already, but you might also be interested in this -- an open-access set of journal papers in Nature Methods, called Visualizing Biological Data, which has lots of different software solutions for generating trees of different sorts, and other things as well.
----
Yes. See this recent post in pharyngula about a paper (Theobald D (2010) A formal test of the theory of universal common ancestry. Nature 465(13):219-222.) that demonstrated this.
It's speculated that there might be life, but not as we know it (not using amino acids of the same handedness, or perhaps not even using organic chemistry), perhaps in some odd crooks and crannies.
----
I assume you probably know about these sites, but if you don't:
http://www.informationisbeautiful.net/
http://www.edwardtufte.com/tufte/
And others, of course.
Hi! Welcome to Pharyngula!
Posted by: DLC
|
June 20, 2010 5:42 PM
Much more informative than that silly "march of time" monkey to ape to ape-man to man thing people put up when they want to insult Darwin and everyone who's come after him. And it has the benefit of being right, unlike DIs "Baraminology" nonsense.
Posted by: waynerobinson4
|
June 20, 2010 6:48 PM
I have just transferred it to my iPad, and I was looking at it in zoomed mode, admiring it, when I noticed that there is a bird with a name of Turdus migratorius. Who would give such a cute American songbird such a name?
Posted by: eternalbookshelf
|
June 20, 2010 9:04 PM
Thank you for posting this, Professor! It's fascinating. I'd always wondered where to find resources such as this.
Posted by: realinterrobang
|
June 21, 2010 12:10 AM
Woahh... It's all 3D and spinny and cool.
I am so tired I feel like I'm going to get sucked into the centre.
...can't sleep...clades will eat me...
Posted by: Your Mighty Overload - Un-True Scotsman
|
June 21, 2010 1:59 AM
Monad @ 70
Ahh, yes, I appreciate there are indeed many more species of insect identified than of protists, however, I suspect that is largely a result of the fact that most protists probably really haven't been characterized.
Posted by: https://www.google.com/accounts/o8/id?id=AItOawnG39uMFt69kwCKZ8DoxtmMCvmzr5chx94
|
June 21, 2010 5:22 AM
Isn't this picture what they call a circular argument? ;-)
Just kidding. Great work!
The Ancestor's Tale is a great book about this sort of stuff. As a non-biologist I loved it.
The diagram looks fractal. Some people here talks about that some groups are under- or over-represented. What does that mean actually? According to what measure? Number of species in each group, number of individuals in each species, total weight in kg, or what?
One interesting thing, the way I understand it, is that it does matter which particular gene, or RNA, or whatever that is inherited, that you are tracing. Y-chromosomal Adam, Mitochondrial Eve, and Most recent common ancestor (for humans) are all different.
Talking about Y-chromosomes, there must be somewhere way back when sex was invented. And that all males of all species are closer related to each other than to any female with respect to these sex-genes.
Posted by: conelrad
|
June 21, 2010 7:42 AM
There is also a version of this about 1.5 meters in diameter mounted on a wall in the science museum in Boston, with a magnifying lens at about eye level so you can zoom in on a small segment of the circumference. As others have noted, very cool.
Posted by: DesertHedgehog
|
June 21, 2010 8:39 AM
Ummm... why does this not show that all evolution is a process aimed at producing Leggy Supermodels?
And why does it leave off the hideous creatures from Elsewhere--- the Evil, Alien Esquimaux, the Vile, Batrachian Manxmen, and the Loathsome, Unhuman Andaman Islanders?
Posted by: KG
|
June 21, 2010 9:26 AM
For an interesting take on the identity of the LUCA (Last Universal Common Ancestor), see On the origin of genomes and cells
within inorganic compartments,
Eugene V. Koonin1 and William Martin. I understand Jack Szostak is following up the ideas in that article, in his work on the origins of life.
Posted by: David Marjanović
|
June 21, 2010 9:44 AM
WTF. How did the turtle* end up in the right place in a molecular phylogeny!?! And in an rDNA dataset at that!?!
<Spock>Fascinating.</Spock>
* There's only one in this tree, the painted turtle, Pseudemys scripta.
You win 1 (one) sniny new Internets. And that already in comment 3. Respect!
*shriek* NONONONONOOOOO!!!
"Most alike" and "most closely related" is not the same thing. Sometimes it is, but by no means always, see comment 34. Species sampling isn't even, and neither are rates of evolution.
All you can do is build an unrooted tree (imagine a tree seen from above instead of a side), using individual similarities rather than the overall percentage of similarity, and then root it by declaring one (or more) of the species the outgroup. You find the outgroup by a priori assuming all other species are more closely related to each other than to the outgroup – an assumption hopefully based on earlier analyses with a different sample of species, or on other evidence.
Incidentally, this means you need at least four species for reconstructing a tree (1 outgroup and 3 ingroup species). There is only one unrooted tree of three (or two or one) species, so it's a trivial case, but there are three unrooted trees of four species.
"Generally". :-D
Nope. Time passes from the center outwards; there is no other axis in the tree. There is no information in it other than that axis and the branching pattern.
Seconded and thirded. I've been thinking this for at least 15 years now.
Actually, the rats + mice are more closely related to the rabbits than to us, as the tree correctly shows.
This plays into the question "what the fuck is a species anyway". As of February 2008, there were 147 different answers out there, all leading to different results when applied, like Mexico having from 101 to 249 endemic bird species – and birds are a bit easier to study than bacteria. I mean, for most species concepts it's enough to just watch them for a bit.
Of course.
All of the above.
(Yes, even by total weight there are more bacteria than trees out there. And that's counting the chloroplasts as part of the trees, not as bacteria.)
<drool>
Posted by: Sven DiMilo
|
June 21, 2010 10:03 AM
Sex was invented long, long before Y chromosomes. Male and female crocodilians, for example, are not distinguishable genetically.
Excuse the SIWOTI syndrome: the painted turtle is Chrysemys picta; the old Pseudemys scripts is the slider turtle, now Trachemys.
Why do you say "Nope" and then retype exactly what I said? All of the tips are extant species, so the total time from the center out is identical for all of them.
No shit. The question was about the placement in the diagram, not the tree topology.
Posted by: David Marjanović
|
June 21, 2010 10:28 AM
Oops.
Sorry. I must have wanted to quote something else than I actually did.
It shouldn't have been. That's my point. :-)
Posted by: https://www.google.com/accounts/o8/id?id=AItOawnG39uMFt69kwCKZ8DoxtmMCvmzr5chx94
|
June 21, 2010 11:09 AM
Really? What is then the difference between male and female crocodilians? Now I am really confused.
Posted by: Erulóra (formerly KOPD)
|
June 21, 2010 11:20 AM
There are differences between them, it's just not genetic.Posted by: John Harshman
|
June 21, 2010 12:27 PM
David Marjanovic:
Posted by: amphiox
|
June 21, 2010 12:28 PM
If I had to hazard a guess, each and every insect species will probably have, on average, about 5 species of "protist" parasites/commensals/symbiotes that are exclusive to it (they're found in no other insect species, anywhere, or anything else for that matter), and maybe 100 species of exclusive bacteria.
Depending on your preferred definition of "species", of course.
The same can be said of each and every species of multicellular lifeform, making the ratio of unicellular to multicellular species 99:1 or greater, not even including any of the free-living unicellular thingies.
It's just that we haven't really looked much at parasites, symbiotes, and commensals yet for most species. Maybe a little bit more in humans thanks to medical interests, but even here, we've only scratched the surface.
Posted by: John Harshman
|
June 21, 2010 12:29 PM
OK, that didn't quite work. Insert /blockquote after the first sentence, would you?
Posted by: Sven DiMilo
|
June 21, 2010 12:32 PM
He's rooting for the paleontologists.
(me too)
Posted by: John Harshman
|
June 21, 2010 12:52 PM
Sven:
You mean *some* of the paleontologists, don't you? Or was Olivier Rieppel excommunicated without me noticing? (References, I presume, in the paper you cite.)
I like paleontologists just fine. Some of my friends are paleontologists. But Gavialis and Tomistoma are sister taxa nevertheless.
Posted by: jeCi KSC
|
June 21, 2010 2:52 PM
A version should be made that includes a timeline in the radial direction showing approximately when different groups diverged from one another =O
Posted by: Ibis3, féministe avec un titre française de fantaisie
|
June 21, 2010 5:16 PM
@57 Thanks for the math. Very cool. Now a very amazing art project would be to install the version with known species (throw in some of the extinct species on the radial lines back to the centre), and use the same map to do a scalar model of the solar system (i.e. so extant species would be at the orbit of Pluto or Eris e.g.) I dunno, just a thought that appeals.
Posted by: Betelgeuse
|
June 22, 2010 12:22 AM
We used the simpler version up above with great success (or so I think) just a week ago!
It was for a display at an educational festival with an underlying theme of biodiversity.
Put it bang in the centre of a poster and surrounded it by lots of neat mini pictures, radially arranged to sort of correspond to the taxa, and voila!
It ended up drawing lots of people who wanted to listen to what everybody had to say, and left us all happy souls who had jibbered-jabbered about science all day :)
Posted by: David Marjanović
|
June 22, 2010 8:22 AM
Senator, I know archosaurs, archosaurs are friends of mine, and turtles, sir, are no archosaurs!
Trying to insert the turtles into the archosaur tree just doesn't work. Not even the aëtosaurs, which are at least fairly heavily armored, share any perceptible number of apomorphies with the turtles. No matter where you put them, you'd need seriously unreasonable numbers of homoplasy.
I'm a student of Michel Laurin. What more is there to say? ;-)
Well, comment 90 is to say. I'm still hoping PZ will blog about it. Read the whole paper, it's open-access (and short).
There is one previous molecular analysis that finds the turtles outside Diapsida, but that's the great big lissamphibian phylogeny by Frost et al. (2006) which had a couple amniotes in it as outgroups, and which used POY (direct optimization), which is a great idea in theory, but has been shown to work badly in practice...
Better yet. Lyson et al. (see comment 90) took his latest dataset (Li et al. 2009, description of Odontochelys), added Eunotosaurus and Proganochelys to it, and the turtles moved out of Diapsida.
BTW, in Rieppel's et aliorum analyses the turtles never came out as archosauromorphs, let alone as archosaurs. They were always lepidosauromorphs, next to either Euryapsida (placodonts and sauropterygians, the latter including the plesiosaurs) or Lepidosauria. Morphological evidence for an archosauromorph position of the turtles is more or less limited to the laterosphenoid, thought to be an archosaur-only bone till some of the authors of the new paper found it in turtles (it's cited in the new one).
Posted by: John Harshman
|
June 22, 2010 11:07 AM
Nevertheless, I'm still holding on to it because it would be so cool if it were true.
I'm unfamiliar with that paper. Could you provide an adequate citation? I presume when you say "a couple amniotes" you mean at least three, there being no other way to make turtles go outside Diapsida. And I've seen examples in which POY actually did improve resolution and accuracy, so it's not completely useless.
Never said or implied otherwise. The archosaur thing is all molecular. I'm waiting for a decent molecular data set, which surely must be in the works in more than one lab, though I don't actually know of any.
Posted by: destlund
|
June 22, 2010 12:40 PM
I'm getting a "phylogenetic tree of me" as a big back piece tattoo with icons of key life experiences instead of species. Nerd tats rock.
Posted by: David Marjanović
|
June 23, 2010 10:11 AM
Here. It's a 370-page monograph. The pdf, accessible from there for free, has almost 10 MB.
Posted by: gmarp84
|
June 23, 2010 4:40 PM
How cool would this be as a crop circle.
Posted by: neuroboros
|
August 4, 2011 7:52 PM
Great blog Professor Myers! And Thanks for the links!