The newest tree of life: many new groups (all bacteria), some cryptic and enigmatic

We’ve known for a while that there are three great divisions in the tree of life: the Eukaryotes (organisms with “true” cells having a membrane-enclosed nucleus, organelles, and a cell membrane); Eubacteria (“true” bacteria); and Archaea (prokaryotes like Eubacteria, but with genes and biochemistry more closely related to those of Eukaryotes than to Eubacteria). Archaea were discovered as “extremophiles”: organisms living in bizarre and difficult habitats like high-temperature springs or vents, and habitats with high concentrations of toxic substances like salt or sulfur; but now we know of many Archaea living in “normal” habitats.

There’s a bit of doubt about the relationship between these three great groups, but most data point to the Archaea being more closely related to Eukaryotes (which include us) than to Eubacteria. That is, we’re more closely related to this:

Than the species above is to this:


E. coli (Eubacteria)

That seems weird, but what happened is that one-celled creatures like primitive bacteria diverged into two groups early in the history of life, and then one of those groups subsequently evolved into the modern Archaea and modern Eukaryotes. In other words, the one-celled ancestor of all complex organisms was also the ancestor of modern Archaea. (The one-celled ancestor was probably similar to living Archaea). That’s not hard to grasp.

A new paper in Nature Microbiology by Laura Hug et al. (reference and free link below) went further, though. Beyond confirming the relationships of the three great branches of life, they showed that the Eubacteria themselves are divided into two great groups that diverged long ago, one of which they call the Candidate Phyla Radiation, or CPR. (Articles on the Hug et al. paper were also written by Carl Zimmer in the April 11 New York Times and by Ed Yong in The Atlantic.)

Further, many of the Archaea and Eubacteria in the Hug et al. paper were not discovered by taking a known organism and determining its DNA sequence. Instead, they were identified by simply isolating DNA fragments from different environments, like salt pans, and sequencing those fragments over and over again until the researchers were satisified that each genome represents a different species. Such organisms can be identified, and in principle reconstructed in the laboratory, but since most of these can’t be cultured in a laboratory environment, we can’t actually see the species. That opens up a whole new way to find species, despite the fact that we can’t see them and may know little about their nature.

The findings also raise the possibility that there are millions of “species” of Eubacteria and Archaea—the species concept becomes a bit fuzzy in these largely asexual groups—that we’ll find in the future. Already, though, it’s clear that Archaea and Eubacteria are the dominant groups on Earth. That’s no surprise since they’ve had the longest time to radiate, and their niches can be everywhere.

The results can be stated briefly. Hug et al. constructed their tree by using the known DNA sequences of 2072 species, and then sequenced the genomes of 1011 new species recovered by sampling DNA from the environment. Total: 3083 species in their trees. The choice of environments for the new species is funny:

This study includes 1,011 organisms from lineages for which genomes were not previously available. The organisms were present in samples collected from a shallow aquifer system, a deep subsurface research site in Japan, a salt crust in the Atacama Desert, grassland meadow soil in northern California, a CO2-rich geyser system, and two dolphin mouths.

Dolphin mouths! But imagine the number of new species they’d get if they’d sampled even more weird environments, like the deep-sea bottom, or even the mouths of llamas. It goes to show how little we know about the diversity of one-celled life.

To construct the phylogenetic tree, the authors used the genomic DNA that codes for ancient proteins present in all three groups: ribosomal proteins. These are the proteins that make up the ribosomes—the little granules in the cell on which proteins are synthesized using as a template the messenger RNA transcribed from the cell’s own DNA.

And here are the trees (you can see the details in the paper, which is available for free):



Figure 1: A current view of the tree of life, encompassing the total diversity represented by sequenced genomes.

In the tree above you see the three great divisions, with the huge radiation of bacteria at the top (including the purple group at upper right, the recently discovered Candidate Phyla Radiation), the Archaea at lower left, more closely related to “true” celled species than to bacteria, and the Eukaryotes, a relatively smallish group in green at lower right. Eukaryotes comprise most of the species we find interesting, but see how small that group  is relative to bacteria-like species. That’s because the ancestors of modern bacteria first appeared about 3.8 billion years ago, but eukaryotes only 1.5 billion years ago. The former have had much more time to radiate.

The groups with red dots are those that lack “isolated” representatives: species that are singleton, distant relatives within the group. I’m not quite sure why they were interested in that, though I am sure a reader will inform me.

Anyway, I give below Hug et al.’s tree based on evolutionary distance (sequence divergence), which gives you a clearer idea of the relationships of the groups. You can see that the Eukaryotes are in fact a small group within the Archaea, so that, at least for these proteins, some Archaea are more closely related to living eukaryotes than to other Archaea. In other words, Archaea is a “paraphyletic” group—one that does not include all the descendants of a common ancestor (some of those descendants are Eukaryotes).


Figure 2: A reformatted view of the tree in Fig. 1 in which each major lineage represents the same amount of evolutionary distance.

One last point. Many of the species in the newly-found CPR must be symbiotic (living in association with other species), because they lack the genes necessary for independent life. As Hug et al. note (my emphasis):

Of particular note is the Candidate Phyla Radiation (CPR), highlighted in purple in Fig. 1 [the first figure above]. Based on information available from hundreds of genomes from genome-resolved metagenomics and single-cell genomics methods to date, all members have relatively small genomes and most have somewhat (if not highly) restricted metabolic capacities. Many are inferred (and some have been shown) to be symbionts. Thus far, all cells lack complete citric acid cycles and respiratory chains and most have limited or no ability to synthesize nucleotides and amino acids. It remains unclear whether these reduced metabolisms are a consequence of superphylum-wide loss of capacities or if these are inherited characteristics that hint at an early metabolic platform for life. If inherited, then adoption of symbiotic lifestyles may have been a later innovation by these organisms once more complex organisms appeared.

The authors call these species “symbiotic,” implying that they get their essential amino acids and nucleotides from other species, but those symbionts can be either mutualistic (both species benefit), commensalistic (the CPR species benefits, there’s no cost to the other one) or parasitic (the CPR species benefits at the expense of its partner[s]). This rely-on-others simplicity, like the CPR radiation itself, was pretty much a surprise. That, along with the vast diversity of species found only by sequencing DNA from the environment, are the two big results of this study.

Sadly, although in principle it’s possible to reconstruct these cryptic bacteria from their genomes—we can synthesize CPR genomes and then inject them into living bacteria whose own DNA has been removed—we’ll never really see them because we can’t culture most of these in the lab. For the time being, and until we develop better culture techniques, we’ll be living in a world full of species whose existence we can discern, but whose characteristics we’ll never see.


Hug, L. A. et al. 2016. A new view of the tree of life. Nature Microbiol.  doi:10.1038/nmicrobiol.2016.48


  1. GM
    Posted April 13, 2016 at 9:37 am | Permalink

    The CPR stuff is absolutely fascinating — there have been some papers in the last few years that suggested something of the sort, but I hadn’t realized so far that we’re talking about such a deep and striking separation.

    Although it should be noted that you don’t get the same picture with rRNA, if we are to believe it, it is amazing that after all these years of sequencing and studies, there are still such surprises…

    • Posted April 13, 2016 at 10:35 am | Permalink

      Yes, really remarkable. I wonder if all that “dark matter” is really symbiotic, or if there are alternative, perhaps more basal metabolic pathways that we have not considered.

      • GM
        Posted April 13, 2016 at 10:37 am | Permalink

        I’m a little suspicious about that too. The future will tell, plenty of interesting topics for research 🙂

        • Posted April 13, 2016 at 12:15 pm | Permalink

          To paraphrase Susan Haack’s line: “It is nice to know we shan’t run out of work!”

  2. Stephen Barnard
    Posted April 13, 2016 at 9:41 am | Permalink

    Good science post. I’m commenting to encourage you to keep posting them! The rapid advances in genetic analysis are astonishing.

    • Posted April 13, 2016 at 10:05 am | Permalink


    • Posted April 13, 2016 at 11:08 am | Permalink


    • eric
      Posted April 13, 2016 at 11:40 am | Permalink

      I second that ‘commenting to keep the posts rolling’!

      • Sastra
        Posted April 13, 2016 at 11:51 am | Permalink


      • Posted April 13, 2016 at 7:46 pm | Permalink

        Me, too! (And I’ve printed and shared the phylogram.)

        • Steven Hill
          Posted April 13, 2016 at 10:54 pm | Permalink

          Fourth. I read but rarely comment.

  3. Posted April 13, 2016 at 9:48 am | Permalink

    Ditto Stephen above! 🙂

  4. Daniel Villar
    Posted April 13, 2016 at 9:50 am | Permalink

    Great article; mind if I ask what makes a bacteria unculturable in a lab? An me what advanced could be done to make these bacteria culturable?

    • GM
      Posted April 13, 2016 at 9:56 am | Permalink

      There is a distinction to be made here, between bacteria that cannot be cultured, and bacteria that have not been cultured. The latter can and not infrequently, do move into the “culturable” category, after people figure out how to grow them.

      But that is often quite difficult — the problem is that in the wild microorganisms exist in communities, in which one species/strain provides various compounds that others than use. Unfortunately, they don’t come with what those things are written in large letters in their genomes, and the biochemical space of possibilities is vast, so it’s not easy to figure out what exactly it is that they need and supply it.

      Then there are intracellular bacteria that are highly adapted to some exotic host, and those are just hopeless when it comes to growing them outside of it.

      • Dominic
        Posted April 13, 2016 at 10:06 am | Permalink

        But surely you can still find the whole organism? As a cell? That does not seem to be the case here…

        • GM
          Posted April 13, 2016 at 10:07 am | Permalink

          Sure you can.

          But how?

          All you have is the sequence.

          People have been using fluorescent hybridization to do that, but those are laborious experiments.

      • Simon Hayward
        Posted April 13, 2016 at 10:12 am | Permalink

        Mostly checking in to confirm that I read the post for PPC(E)s benefit!

        However, (since I’m here and know almost nothing about life at this level) it would seem that these “crippled” genomes and the communities that GM mentions represent a selection pressure for the sort of endosymbiosis that likely resulted in eukaryotes. That being said, eukaryotes are apparently monophyletic, and you have to wonder why this event was so unlikely as to only successfully occur once. Anyone have any thoughts on this?

        • Posted April 13, 2016 at 1:43 pm | Permalink

          I – and many others – have always wondered why the evolution of a prokaryote to the first eukaryote required more time than abiogenesis (i.e. the “chemical evolution” of the first prokaryotes from non-living matter).

          • loren russell
            Posted April 13, 2016 at 5:04 pm | Permalink

            Why were eukaryotes so late on the scene? I’d guess that the answer is that there was a huge adaptive valley from prokaryotic “sex” and reproduction, which works well enough, to the intricacies of intracellular and nuclear membranes. You really had to select for big cells to get there, when biofilms got it done well enough.

            Even then, there wasn’t a necessary path to more complex life — another billion years passed before you got to multicellular organisms.

          • Gregory Kusnick
            Posted April 14, 2016 at 1:50 am | Permalink

            It seems likely that environmental constraints played a significant role in the timing. Mitochondria implement aerobic respiration, right? So there’s no point having them until the atmosphere has been sufficiently oxygenated to make aerobic respiration profitable.

  5. Merilee
    Posted April 13, 2016 at 9:50 am | Permalink


  6. Kevin
    Posted April 13, 2016 at 9:52 am | Permalink

    Wow. Two things struck me.

    One, life is much more connected than most people are aware.

    Two, we focus so much of our attention on ‘big’ life, i.e., mostly human stuff, that we lose sight that most life is just about moving information around with as little energy as possible. And this frequently leads to massive speciation on scales, I think, we are still only beginning to fully appreciate.

  7. steve oberski
    Posted April 13, 2016 at 9:52 am | Permalink

    According to Douglas Adams, dolphins are the 2nd most intelligent species on the planet, so why not dolphins ?

    Humans are bipedal creatures from Earth, and the third most intelligent species on that planet, surpassed only by mice and dolphins. Originally thought to have evolved from proto-apes, humans may in fact be descendants of Golgafrinchan telephone sanitizers, account executives, and marketing analysts who were tricked out of leaving their home planet to arrive on the planet ca. two million BC.

  8. rickflick
    Posted April 13, 2016 at 9:54 am | Permalink

    So, I’m wondering if the symbiont forms live inside larger organisms – like gut bacteria? Or do they live in close proximity so they can absorb nutrients?

    • Dominic
      Posted April 13, 2016 at 9:56 am | Permalink

      It highlights the colonial nature of life. Bodies are co-operatives of cells that cannot survive on their own.

    • GM
      Posted April 13, 2016 at 9:58 am | Permalink

      Most of those come from environmental samples, so they are likely symbionts with other microorganisms — unicellular eukaryotes or small invertebrates.

      BTW, both of those groups represent vast understudied universes of diversity on their own.

    • Posted April 13, 2016 at 10:48 am | Permalink

      When I was taught biology, all symbionts were mutualistic.

      Has there been a shift in usage or has there always been variation among definitions?


      • GM
        Posted April 13, 2016 at 10:52 am | Permalink

        There is a narrow and broad definition of “symbiosis”. The broad one is more proper, because there is no sharp distinction between mutualism and everything else, there is a smooth continuum from real parasitism to true mutualism, all the way in between.

      • Diane G.
        Posted April 16, 2016 at 12:33 am | Permalink

        In short, the answer is “yes.” Drove me crazy when I first heard of it.

  9. Dominic
    Posted April 13, 2016 at 9:55 am | Permalink

    Not sure I understand this –
    “since most of these can’t be cultured in a laboratory environment, we can’t actually see the species.”
    are you saying that they CANNOT find these organisms living, or just did not? Is this like sampling pollen & then trying to reconstruct the plants they are from?

    By the way, ancient canid from Siberia…!

    • darrelle
      Posted April 13, 2016 at 10:35 am | Permalink

      I think they mean something like “without a tremendous effort beyond way beyond our means.” Their methods resulted in them looking at large varieties of genetic material, but not actual organisms.

      In principle they (or anyone) could go find the actual organisms, but to find even just one could be a much, much larger task than merely collecting “gunk” from an interesting site and running a genetic analysis on it which yields genetic material from dozens or hundreds of different species.

      I recall reading about a study some years ago, but after the advent of easy genetic sequencing, where researchers more or less randomly collected water samples from the ocean and sequenced the genetic materials in the samples. They found oodles (technical term) of genetic material from oodles of different species, the large majority of which (something like 90% +), they were surprised to find, there was no previous record of.

  10. Craw
    Posted April 13, 2016 at 9:55 am | Permalink

    I am gonna read this.

  11. Joseph McClain
    Posted April 13, 2016 at 10:25 am | Permalink

    When I first saw this “tree,” I started searching for Eukaryotes (sue me for kinship preference) near the top. That’s where we usually are! Anyone else?

    • Posted April 13, 2016 at 10:50 am | Permalink

      The typical expectation of a privileged white male eukaryote. Sensitivity training for you.

    • James Walker
      Posted April 13, 2016 at 12:11 pm | Permalink

      Check your eukaryote privilege!


      • Joseph McClain
        Posted April 13, 2016 at 3:19 pm | Permalink

        “Privilege?” It’s no privilege! It’s a prison! Why do you think there are all of these cells!

  12. Posted April 13, 2016 at 10:26 am | Permalink

    Thanks for posting this. I am presently reading about the evolution of the complex cell in Nick Lane’s book, and this fits right in.

    I wonder whether a group of symbiont “species” in the CPR unable to survive independently could be viewed as a single organism.

    • Mark Sturtevant
      Posted April 13, 2016 at 12:42 pm | Permalink

      I am working thru the same book, as time allows. He gets into the point that although there are occasional prokaryotes with endosymbionts, it was not very transformative in those cases as it was when the ancestor to the eukaryotes picked up endosymbionts that later became mitochondria. His point was that in our case all hell broke loose because this allowed eukaryotes to become flush with ATP production. But why did it not do so in the other cases?

      • Posted April 13, 2016 at 1:21 pm | Permalink

        Which book?

        • Posted April 13, 2016 at 2:05 pm | Permalink

          Life Ascending

          • Mark Sturtevant
            Posted April 13, 2016 at 2:22 pm | Permalink

            Also (and more recently) The Vital Question. He goes into a lot more depth there.

            • Posted April 13, 2016 at 2:36 pm | Permalink

              I will read that if I ever get finished Life Ascending. I think I am already at my depth.

              • Posted April 13, 2016 at 4:23 pm | Permalink

                Thanks both of you. I already have both on Kindle. Must get reading them …

            • Sarr
              Posted April 14, 2016 at 3:34 am | Permalink

              I remember Life Ascending being a fascinating read, but I hadn’t heard of The Vital Question before. I’ll have to check it out. Thanks for the tip! 🙂

              • Packard Baggins
                Posted April 14, 2016 at 7:00 am | Permalink

                I may be mistaken, but my takeaway from The Vital Question is that the split between Eubacteria & Archaea (in Dr. Lane’s view) might have come even before these organisms became free-living entities–that is, while they were still bags of chemical reactions occurring in those white smoker chimneys on the ocean bottom.

  13. Posted April 13, 2016 at 10:27 am | Permalink

    Fascinating stuff. I would never have guessed at the size of the CPR or its symbiotic majority. Knowing that now though, I guess it makes intuitive sense that such a large subset of life would try to piggyback off other life. Why make your own amino acids when you can use somebody else’s?

  14. Mark Sturtevant
    Posted April 13, 2016 at 10:30 am | Permalink

    I had been meaning to pick up on this story, and this effort made my life a bit easier!

  15. Posted April 13, 2016 at 10:39 am | Permalink

    I wonder if the authors are using “isolated” to mean “isolated in a lab and cultured” rather than isolated in a phylogenetic sense. That might explain their interest in it.

  16. Posted April 13, 2016 at 10:45 am | Permalink


    IN case anyone was wondering, we’re somewhere among the opisthokonta at the bottom right, along with all other animals and fungi!


    • Posted April 13, 2016 at 7:50 pm | Permalink

      Thanks! I was trying to find us. 🙂

      (Still ‘leafing’ my way through this substantial post plus comments, PCC(E).)

    • HaggisForBrains
      Posted April 14, 2016 at 5:54 am | Permalink

      Thanks from me as well. Did anyone else notice the two Norse gods lurking among the eukaryotes? Oops, I see Eric did, further down.

      • Posted April 14, 2016 at 6:02 am | Permalink

        I really wanted to see a “You are here” arrow on this. (Which I forgot to add when I posted this on Facebook.)


  17. Posted April 13, 2016 at 10:50 am | Permalink

    YAY Science!

    I’m beginning to understand the difference between those three categories.

    Thanks PCC(e)!

  18. Steve Gerrard
    Posted April 13, 2016 at 10:58 am | Permalink

    So very, very, not intuitive. I don’t know what to make of the tree when it is laid out like that. It seems alien. We seem to know a whole lot about one small branch of it, and not that much about all the rest of it.

    • Posted April 13, 2016 at 1:20 pm | Permalink

      To me, it is a “pale blue dot” moment.

      More or less all the life you have ever seen or otherwise experienced in your life (aside from infections, gut flora, etc.) all animals, almost all plants, to say nothing about finer divisions like vertebrates, mammals, primates, etc.) are ALL in a little side branch of the great family of archaea.

      This is humbling and perspective-giving to me.

      • Diane G.
        Posted April 16, 2016 at 12:48 am | Permalink

        Beautifully put!

  19. Posted April 13, 2016 at 10:58 am | Permalink

    Prof. Coyne–
    I am leaving this comment to let you know that I read and loved this! I will go and read at least Carl Zimmer’s or Ed Yong’s article and maybe even the article itself. I AM READING YOUR SCIENCE POSTS!!!!!

    • Dominic
      Posted April 13, 2016 at 11:15 am | Permalink

      PCC[E] is a kitten & you have just tickled his ears 🙂

    • Posted April 13, 2016 at 11:29 am | Permalink

      Your science posts are read and they have impact!

      I have read your post and I will read the paper as well as the take(s) of Ed. Yong & Carl Zimmer.

      I will USE the Information for the preparation of a Cell Biology Class.

      For a discussion of the relation of Eubacteria, Archea and Eukaryota (and the
      Three Domain Hypothesis) I further recommend an article series on Larry Morans Blog Sandwalk.

      Thanks your Website!

      P.S. Your posts on Group Selection sparked a (not closely finished) reading session ranging from statistics to population genetics.

  20. Posted April 13, 2016 at 11:13 am | Permalink

    Sorry, this is not my first science. Do I understand correctly, if simply, that these symbionts cannot make proteins because they don’t have the energy for it, lacking aerobic respiration (no Krebs cycle and so forth)? And that is why they must be symbionts? Thanks for any help.

    • chascpeterson
      Posted April 13, 2016 at 11:55 am | Permalink

      No. They can make ATP for their energy needs, but with use of anaerobic pathways instead of aerobic (Krebs cycle, respiratory chain). And they can make proteins from amino acids (that’s what ribosomes do), and nucleic acids from nucleotides, but they lack the ability to synthesize their own building blocks (amino acids and nucleotides). So they must be co-opting somebody else’s biochemistry to obtain those building blocks.

      • Posted April 13, 2016 at 12:23 pm | Permalink

        So why can’t they make nucleotides or amino acids? I understand we get amino acids from digesting the proteins we eat. Where do bacteria get them? THanks.

        • reginaldselkirk
          Posted April 13, 2016 at 12:36 pm | Permalink

          They steal them from somebody who already has them.

        • Mark Sturtevant
          Posted April 13, 2016 at 12:46 pm | Permalink

          Parasites and other symbionts often lose features b/c they are supplied, willingly or otherwise, from their hosts. It takes energy to build and maintain genes, and that energy could be used for other things. So in this case bacteria are shedding genes they dont’ need b/c it gives them a growth advantage over the ones who don’t shed the genes.

  21. squidmaster
    Posted April 13, 2016 at 11:20 am | Permalink

    This is one of the most fascinating articles to come along in a while. I was also fascinated by the CPR. It’s easy to see why these organisms were not recognized earlier. They surely don’t live independently and almost certainly don’t cause human disease. Neither, with a few exceptions, do Archaea. Even if one looked at those dolphin mouth scrapings under a microscope, it would be impossible to identify (with maybe a few exceptions) individual organisms. There may be very few members of each type of beast and they all look pretty much alike.

    One point that I don’t quite understand and perhaps someone more familiar with the technical side of these methods could illuminate the issue. The pretty figure was constructed by comparing nucleotide sequences that coded for ribosomal proteins. This analysis placed Eukarya within the Archaea. If one looks at smaller subunit ribosomal RNA, OTOH, one derives a tree that makes Eukarya a sister taxon of Archaea. My questions are as follows. Is it the case that these findings represent two mutually exclusive interpretations; i.e., Eukaryotes arose within the Archaea or Eukaryotes arose as a sister taxon to the archaea. Is there any way for both of the experimental findings to be true? I can’t see any way around the latter, but the alternative is that one of the genomic analyses is flawed. Any light shed would be mostly appreciated.

    • GM
      Posted April 13, 2016 at 11:27 am | Permalink

      1) It was done on protein sequences, not on nucleotide ones. 16 ribosomal proteins to be exact. Which is a little contentious — to what extent to these 16 proteins represent the true phylogeny? But they’re doing their best so it is what it is

      2) Eukaryotes indeed arose from within Archaea, that has at this point become almost a consensus. The larger and more modern phylogenies, built on multiple genes and a deep sampling of archaeal diversity with more sophisticated algorithms all return those kind of trees. Eukarya sister to Archaea was recovered in the past because those things were not available.

      • reginaldselkirk
        Posted April 13, 2016 at 1:06 pm | Permalink

        1) It was done on protein sequences, not on nucleotide ones. 16 ribosomal proteins to be exact. Which is a little contentious — to what extent to these 16 proteins represent the true phylogeny?

        They are using the gene sequence which codes for the protein, not the peptide sequence.

        As to what represents the “true phylogeny” that is a difficult question, which has been asked before, and I suspect that there is no definitive answer (although some answers may be better than others)

        I would consider the ribosomal RNA to be more “core” than the ribosomal proteins; this is because I support the “RNA World” scenario.

        But, they have their criteria, and should apply them consistently and well. I expect that in most cases the results will be the same.

        It occurs to me that if an organism had multiple chromosomes, their approach could not assemble complete genomes of that organism. We can do it for eukaryotes because we can isolate examples of the organism and associate the chromosomes with that organism.

      • squidmaster
        Posted April 13, 2016 at 1:39 pm | Permalink

        Thanks to GM and reginald (RS) for elucidation. My reading is the same as RS; they used the DNA sequences for the ribosomal proteins, not amino acid sequence. And, from the same reconstructed genomes, they used the sequences for the SSU rRNA. The authors found, though, that the ribosomal protein sequences and rRNA sequences arrived at different trees. I suppose they are using ribosomal genes because of low rates of mutation and drift. The authors clearly believe that the tree showing Eukarya arising from Archaea is more likely, given the prominence accorded to that tree (and *much* prettier graphics). The sister eukaryote tree was stuck in the supplementary material. is it, as GM suggests, that eukaryotes as Archaea is widely viewed as true or is the ribosomal protein gene data more compelling than the rRNA data? This may be a bit techinical, so a reference would be fine.

        • VJ
          Posted April 14, 2016 at 7:49 pm | Permalink

          It’s not so clear in the text, but from a close reading of the methods they used the predicted amino acid sequences of these proteins, based on the nucleotide sequencing data. The clincher is Supplementary Dataset 1, which shows the amino acid alignments they used to build the tree.

    • Posted April 13, 2016 at 11:48 am | Permalink

      To clarify the issue I recommend to
      have a look at Larry Morans take:

      ‘The origin of eukaryotes and the ring of life’ at ‘Sandwalk’

      • Mark Sturtevant
        Posted April 13, 2016 at 12:51 pm | Permalink

        This is a point I was hoping to see somewhere here. Getting the above tree where eukarya emerges from within the archaea depends on using genes that mediate gene expression. But if one uses genes for metabolism, like genes for electron transport proteins, the eukarya come out closer to the eubacteria (!) Hence the view from the ‘Ring of Life’ hypothesis — that we eukaryotes are a kind of hybrid born of fusion between an archaeon and eubacterial cell.

        • Posted April 13, 2016 at 2:00 pm | Permalink

          I guess, eukaryotic genes for electron transport proteins actually come from eubacterial endosymbionts (mitochondria), even if they reside in the nucleus today.

          • Mark Sturtevant
            Posted April 13, 2016 at 2:27 pm | Permalink

            Yes, but in the ring of life scenario the eubacteria did not enter an ancient eukaryote cell, but entered an archaeon cell. The combined cell then evolved –> the eukaryote. The internalized bacterium migrated a lot of its DNA to the archaeon genome, and the archaeon genome lost its genes that were made redundant.

        • Torbjörn Larsson
          Posted April 13, 2016 at 2:48 pm | Permalink

          I guess it is a sort of selfish gene perspective, where heredity ability trump metabolic ability. The lineages are best traced by rRNA et cetera.

          • Torbjörn Larsson
            Posted April 13, 2016 at 2:50 pm | Permalink

            I meant to say the those (like me) that doesn’t heave to a ring model has a sort of most useful tracer perspective.

          • reginaldselkirk
            Posted April 13, 2016 at 3:33 pm | Permalink

            Whatever technique you’re using, do it consistently and as well as possible.

            It is an interesting and informative result that the ribosomal genes and the metabolic genes yield different results. Argumentation over which is “better” is like arguing over the superiority of tracing patriarchal vs. matriarchal ancestry.

  22. Posted April 13, 2016 at 11:29 am | Permalink

    All those red dots!

    It’s amazing how much terrain can open up through the data from a new technique.

  23. DTaylor
    Posted April 13, 2016 at 11:31 am | Permalink

    This was fascinating. I’ve been intrigued and confused for some years about the relationships between these very early life forms. I’m still intrigued and confused, but now I don’t feel so bad about it. I found the “evolutionary distance” graphic very helpful, never mind the shock of seeing the Eukaryotes embedded in Archaea. Great post. I’ve flagged it for later close reading. Big thanks!

  24. Michael
    Posted April 13, 2016 at 11:44 am | Permalink

    I’m sure everyone has heard the old joke; A biologist was walking on the beach when he found an old lamp. He picked it up and a genie appeared. “I will grant you one wish.” The biologist said “I thought it was three.” The genie said, “Well since the recession it’s down to one.” The biologist said, “I always wanted to visit Hawaii, but I afraid to fly, and get very sea sick on boats, could you make me a highway from the west coast to there?” The genie responded, “Gee, that’s really hard, could you make another wish.” “Okay, could you give me the definition of species?” The genie thought for a moment, then asked, “Do you want a two lane or a four lane highway?”

    • Mark Sturtevant
      Posted April 13, 2016 at 12:53 pm | Permalink

      Consider it stolen! 🙂

    • Posted April 13, 2016 at 7:58 pm | Permalink

      I like that one! Yes!

    • Diane G.
      Posted April 16, 2016 at 12:55 am | Permalink


  25. Denis Westphalen
    Posted April 13, 2016 at 11:58 am | Permalink

    Dumb question from a non-biologist: are there multicellular organisms outside the Eukariotes group?

    • Posted April 13, 2016 at 12:09 pm | Permalink

      Generally, no, but there are some prokaryotes that form aggregations that may have a slight division of labor. See here.

    • JohnW
      Posted April 13, 2016 at 12:13 pm | Permalink

      Yes and a good question![]

      • Denis Westphalen
        Posted April 13, 2016 at 3:37 pm | Permalink

        Thanks for the reply and links.

  26. JohnW
    Posted April 13, 2016 at 12:05 pm | Permalink

    From the methods: “The full tree inference required 3,840 computational hours on the CIPRES supercomputer.”

    Whoa…. First thing I thought of was sample integrity and contamination criteria, but seems they were pretty stringent and discarded many partial genomes. Metagenomic assembly is truly mind boggling…..Thanks for posting.

  27. Gregory Kusnick
    Posted April 13, 2016 at 12:08 pm | Permalink

    In other words, the one-celled ancestor of all complex organisms was also the ancestor of modern Archaea.

    Perhaps I’m misreading it, but this seems backwards to me. The common ancestor of Archaea is also a common ancestor of Eukaryotes. But surely there’s a more recent common ancestor of Eukaryotes that’s not a common ancestor of Archaea.

    • Posted April 13, 2016 at 1:13 pm | Permalink

      Yes, I think that has to be correct.

      We can say that all archaea and all eukaryotes share a common ancestor (which was not the one that split from the (other) archaea and evolved into all eukaryotes).

      • Gregory Kusnick
        Posted April 13, 2016 at 7:32 pm | Permalink

        Another possible reading is that there exists some living descendant of the most recent common ancestor of Eukaryotes that is not itself a Eukaryote. While logically possible, I suppose, this seems rather unlikely, and certainly not a given.

  28. pck
    Posted April 13, 2016 at 12:10 pm | Permalink

    I wonder if the use of a single genomic region for their phylogenetic reconstruction could pose some problems. Normally you would want to use multiple unlinked markers to avoid things such as horizontal gene transfer messing up your phylogeny. Not sure how prone ribosomal proteins are to that (16srRNA is said to be comparatively immune to transfer). Horizontal transfer of that region between organisms from the same weird environment could make them appear more closely related than they are. It’s an impressive study and I’d like to see it repeated with better markers for just bacteria to see if the CPR branch holds up (chances are it’s already been done and I’m not up to date with the literature)

    This comment was made in support of the Keep Up The Science Posts Angry Cat Man initiative.

    • Posted April 13, 2016 at 12:32 pm | Permalink

      « This comment was made in support of the Keep Up The Science Posts Angry Cat Man initiative. »


      • Mark Sturtevant
        Posted April 13, 2016 at 2:29 pm | Permalink

        I belong to the SPotACMS: Science Posts of the Angry Cat Man Society.

  29. alnitak
    Posted April 13, 2016 at 12:12 pm | Permalink

    We’ve come a long way from Bergey’s Manual, which itself presented a bewildering array of bacteria. The role of bacteria in our lives is still underappreciated, and an analysis like this shows how much more there is to know. They should try to analyze all the genomes on people!

    • Barbara Radcliffe
      Posted April 13, 2016 at 5:04 pm | Permalink

      Yes, microbiology has come a long way since I did my degree in the sixties. I still have my vintage Bergey’s Manual which I keep as a souvenir.

  30. Posted April 13, 2016 at 12:18 pm | Permalink

    Wow. I am reminded of those diagrams at the end of _The Ancestor’s Tale_ which shows what a twig creatures like we are on the tree of life. This confirms it!

  31. Charlie Jones
    Posted April 13, 2016 at 12:26 pm | Permalink

    I was stunned by the huge ancient split between the eubacteria and the “candidate phyla radiation”. Is this grouping totally new or am I an ignorant non-biologist?

    I was also very surprised to see eukaryotes nested within the archaea, instead of a branch away from all archaea. I wonder how likely this is to hold up with future studies using different analytical data.

    FINALLY, I looked up the timing of the split between eukaryotes (Homo sapiens) and an archaea near the big split (Korarchaeum cryptofilum) on The age is about 4.3 billion years ago, which is about 500 million years before the earliest hint of life preserved in the rock record. That is an ancient split!

    • Torbjörn Larsson
      Posted April 13, 2016 at 3:07 pm | Permalink

      The branching of eukaryotes within archaea got its first more tangible support. I think, with Spang’s discovery of Lokiarchaeota and its eukaryote like toolkit for phagocytosis (possibly, not confirmed). [ ] So I note that the new tree more or less confirms that. (But now there is a more and closer lineages within that clade.)

      I think your split date is way too early. As Jerry notes, the split is believed to be much younger, from finds of eukaryote metabolic trace fossils and microfossils.

      But we can go along your analysis too, to see it is too early. Valley’s latest zircon review puts a habitable Hadean ocean > 4.3 Ga, while TimeTree dates the first split between Bacteria and Archaea > 4,2 Ga.

      One can argue about what hints of early life means, but there is a (rather two) fossil candidates that dates > 4.1 Ga thanks to surviving zircons, consistent with the above dates. [ ]

      It lacks the necessary replication and context so far, but despite lacking the usual context possible abiotic pathways have been eliminated according to the peer reviewed paper. (The first author Bell has since given a SETI Talk, and claims they have an undated – possibly too young – replication under analysis.)

  32. Scientist
    Posted April 13, 2016 at 12:52 pm | Permalink

    Great post and discussion. These PCR methods also show numerous unknowns among the protists, those eukaryotes not placed in fungi, plants or animals. Though some show affinity with known groups, many, perhaps most, have no close relatives. The true diversity of life has yet to be described.

    • Mark Sturtevant
      Posted April 13, 2016 at 1:03 pm | Permalink

      That is another interesting subject, showing that the ‘protists’ form a huge and diverse tree, with the so-called animal, fungi, and plant kingdoms being but twigs on the great protistan tree.
      Protists are us.

  33. Posted April 13, 2016 at 12:59 pm | Permalink

    This is amazing stuff. I particularly like how the lower (linear) tree of life really puts us eukaryotes in our place relative the rest of the living world.

    As I tell people often, to a first approximation, all life on earth is bacteria and archaea (by mass, number of species, and diversity).

    • Posted April 13, 2016 at 2:06 pm | Permalink

      “As I tell people often, to a first approximation, all life on earth is bacteria and archaea.”
      Some say it may become, if people do not suddenly become responsible (though eukaryotes have had hard times in the past and have endured – some of them at least).

    • Posted April 14, 2016 at 10:10 am | Permalink

      I forgot to list as a measure: Number of individual organisms.

    • Gregory Kusnick
      Posted April 14, 2016 at 12:59 pm | Permalink

      For what it’s worth, Wikipedia says that total eukaryotic biomass rivals that of prokaryotes.

    • Diane G.
      Posted April 16, 2016 at 1:07 am | Permalink

      “…really puts us eukaryotes in our place relative the rest of the living world.”

      Good thing our particular species of eukaryote has dominion over all the rest! 😉

  34. Posted April 13, 2016 at 1:02 pm | Permalink

    I wonder if these “symbionts” might be the remnants of the initial steps of evolution, reticulate evolution where they actually live as communities and share via gene and other transfer using as central protocol their shared generic code. Since they seem rather separate from the rest, they might even have a different code!

    Very cool stuff

    • Posted April 13, 2016 at 2:30 pm | Permalink

      This was actually thought of the intestinal parasite Giardia lamblia, which has no mitochondria and seemed to have branched off from other eukaryotes very early. However, it turned out that the lack of mitochondria was secondary, and the impression of long independent evolution was a misinterpretation of Giardia’s high mutation rate; the latter, in turn, was due to the high body temperature of the mammalian host.
      Personally, I think that within a highly evolved host, there is little chance to find a symbiont keeping primitive traits.

    • reginaldselkirk
      Posted April 13, 2016 at 3:39 pm | Permalink

      Since they seem rather separate from the rest, they might even have a different code!

      No, at least not the subjects of this paper. They used genes coding for ribosomal proteins, so use of the familiar genetic code is assumed.

  35. Pascal Nelson
    Posted April 13, 2016 at 1:35 pm | Permalink

    Read and appreciated. Keep ‘m coming.

  36. eric
    Posted April 13, 2016 at 2:10 pm | Permalink

    Hat tip to Hemant Mehta who first noticed it, but note the presence of Loki and Thor just to the left of the Eurkaryotes. Is that a mischievous addition to the tree by the authors, or a real branch that reflects mischievous naming of two real organisms by prior biologists?

  37. Torbjörn Larsson
    Posted April 13, 2016 at 2:22 pm | Permalink

    Thanks for this, due to my recent illness I haven’t had time to read the article, and I definitely didn’t know about (or connect to, since CPR species has been known for a while) the interesting implication of early incomplete metabolic cycles!

    That reminds me of hypotheses of early virus lineages which split off early and simplified, perhaps because they too were incomplete but evolved aggressive parasitism. But it could also tell us of early Hadean conditions.

    I note that Spang’s discovery of Lokiarchaea as sister lineage to Eukaryota is close to correct, but a clade with an older split of Thorarchaeota has now taken its role. The new lineages and an earlier split than Lokiarchaeota should help elucidating how eukaryotes evolved. Lokiarchaeota had a phagocytosis toolkit (but it is unknown if it can do it), but Thorarchaeota may be less well prepared for an endosymbiosis.

    • Torbjörn Larsson
      Posted April 13, 2016 at 2:24 pm | Permalink

      Oh, ninja’ed by eric!

  38. Paul S
    Posted April 13, 2016 at 2:33 pm | Permalink

    Cool, and this was much easier to understand than the post about the Templeton grant to study non-evolutionary based evolution.
    I don’t understand it all, but it makes me feel less stupid.

  39. rjcatt
    Posted April 13, 2016 at 3:05 pm | Permalink

    Fascinating post and good discussion — cannot add anything, really, but want to do my bit to encourage more science posts by PPC(E)

  40. Vaal
    Posted April 13, 2016 at 3:09 pm | Permalink

    Fascinating post. Though I think I’ll have to read it again to absorb it all.
    I remember debating with some creationists at one point and I brought in examples that appealed to Phylogenomic evidence. There was a new pathogen being studied, which was worrying because, among other things, there was no known treatment and it did not respond to standard chlorine treatments in water, making it not only a natural threat, but a potential water born bioterrorism agent. In particular, although it was malaria-like, unlike malaria it didn’t have an apicoplast organelle to target with drug therapies.

    It was by employing common descent, “tree of life” Phylogenomics that allowed researchers to pinpoint ancient intracellular and horizontal gene transfer points. They found genes transferred from distant phylogenetic sources, which would be targets for new drug therapies (and their phylogenetic distance from our own metabolic pathways would help target the pathogen, not us).

    Having described (in much greater detail) the detective story undertaken, guided by a common descent theory, with successful predictions and results, I asked what other methods would creationists have used to get successful results. None were forthcoming of course, and the handwaving reached tornado-producing proportions.

    (As a layman, all that stuff strains the limits of my understanding in biology, so posts like today’s are always welcome information).

  41. Posted April 13, 2016 at 3:13 pm | Permalink

    As a white eukaryote checking her privilege :-), I’d respectfully disagree with the description of eukaryotes as “organisms with “true” cells”, though it may be helpful to lay readers.
    I think that the cell is routinely taught wrong, because it is introduced using the example of eukaryotic (either animal/human or green plant) cell. It is too complex to be understood easily, and looking at it, it is difficult to be convinced that life has evolved from non-living matter.
    Therefore, I am always trying to introduce prokaryotic cell as the cell and then to describe eukaryotic cell as an upgrade.
    Here is a drawing intended to introduce the cell for 1st time to middle school kids:
    The text is in a language better than Ehglish :-), which unfortunately makes it unintelligible. The legend is: cell membrane, cytoplasm, chromosome. Because the target audience is too young for the concept of protein synthesis, there are no ribosomes in the cytoplasm. In a later text, I explain that this drawing is over-simplified to the point of being wrong, because only a prokaryotic cell could be that simple, but it would have a circular chromosome.

    • Diane G.
      Posted April 16, 2016 at 1:20 am | Permalink

      With a faint memory of the cyrillic alphabet I was able to get “membrane,” “cytoplasm,” and “chromosome” with no problem. Gotta love cognates!

      • Posted April 16, 2016 at 8:25 am | Permalink

        Thank you for visiting my weird corner!

  42. µ
    Posted April 13, 2016 at 4:25 pm | Permalink

    Good post. Better than Zimmer’s writeup in NY Times (which was also excellent). Maybe consider becoming a science writer for NY Times, get paid for your posts?

    Not sure I agree with “largely asexual groups”. All kinds of sexual and parasexual exchange happening that only evolutionary biologists are not afraid to ask about.

  43. Posted April 13, 2016 at 5:04 pm | Permalink

    Great post!

    Archaea reconfirmed as paraphyletic, not surprising. We are a specialised group of Archaea.

    The other thing that I find interesting is that there is a ‘morphologically’ very meaningful tree, because we are being told so often that there is no tree of life to procaryotes as they constantly exchange their genes. Quite apart form that being incoherent because people conceptually mix up systematics-relevant lines of descent with systematics-irrelevant horizontal gene transfer it appears that one can get an idea of relationships if using the right genes. I mean, otherwise there should be what is chemically and ecologically an Achaea nested in the Eubacteria or vice versa.

  44. Neil Taylor
    Posted April 13, 2016 at 6:23 pm | Permalink

    Great science post.

    With Eukaryotes paraphyletic it really starts to demolish the idea of the three domains of life.

    One question for anyone who knows – with Eukaryotes the relationship between organisms can be confused if you only use one part of the genome – say the mitochondria – to determine the closeness of a relationship.

    Might this study have a similar problem and later studies show the paraphyletic relationship is an artefact and not representative of the true genetic relationship.

  45. Posted April 13, 2016 at 6:59 pm | Permalink

    I just wanted to comment to say that this is one of those very fascinating science posts that I read, but never click to comment because I really don’t have the knowledge to contribute to the conversation. I do hope you continue posting them though 🙂

  46. Posted April 13, 2016 at 7:31 pm | Permalink

    Very interesting article. Is there any evidence that mutualistic relationships occur more often than commensalistic relationships, and that commensalistic relationships occur more often than parasitic relationships? I would expect that parasites generally put pressure on their host species, which in turn compromises the likelihood of their own survival. The opposite should be true for an organism the enjoys a mutualistic relationship with another organism.

  47. George
    Posted April 13, 2016 at 8:41 pm | Permalink

    The figure 1 chart is awesome – visually. I need to study it a bit more though.

  48. Dale Franzwa
    Posted April 13, 2016 at 11:16 pm | Permalink

    Terrific post, Jerry. I can hardly wait for the Reader’s Beefs to show up telling you how all this PROVES! evolution doesn’t work and that GODDIDIT but not explaining how he didit.

  49. dave
    Posted April 13, 2016 at 11:23 pm | Permalink

    They are just called ‘bacteria’ not ‘eubacteria’

  50. Posted April 14, 2016 at 12:05 am | Permalink

    Awesome post!! Far, far better than Carl Zimmer’s article in NYT I read a couple of days back (No offence, Carl). Please keep writing!

  51. Terryl Stacy
    Posted April 14, 2016 at 9:32 am | Permalink

    Just to let you know, I read this. I read all of your science posts. Keep up the good work.

  52. Posted April 14, 2016 at 6:49 pm | Permalink

    It is simply amazing that a handful of nucleotides can produce such variety. What a fabulous chart!

  53. Diane G.
    Posted April 16, 2016 at 1:30 am | Permalink

    This is beyond fascinating! PCC is the best science writer going, IMO–able to put complex subjects into a form not only understandable but literally enjoyable to read.

    Not, of course, that I’m grasping all the intricacies and details/qualifications that the brilliant commenters here are posting, but this article was a great foot in the door!

    “I have no doubt that in reality the future will be vastly more surprising than anything I can imagine. Now my own suspicion is that the Universe is not only queerer than we suppose, but queerer than we can suppose.”

    –J. B. S. Haldane

One Trackback/Pingback

  1. […] [2]… […]

%d bloggers like this: