More Recent Comments

Friday, May 08, 2015

Ford Doolittle talks about the tree of life

There's an interview with Ford Doolittle in PLoS Genetics [The Philosophical Approach: An Interview with Ford Doolittle].

Ford has lots of things to say about the origin of introns, the tree of life, transpsosons, ENCODE, and the meaning of "function." Here's the bit about the tree of life.
Doolittle:I’m drifting off into philosophy in my old age. People have gotten quite heated over whether or not there is a tree of life. And really the question is what do you mean by the “tree of life?” And clearly—this is sort of Eugene Koonin’s position, which he calls the “forest of life”—there is a consensus signal.

If you make trees of all genes, it’s not completely random. Patterns emerge. And then you think, “Well duh, what the hell else do you expect?” Nobody would expect that lateral gene transfer wouldn’t respect geography or biochemistry or ecology or whatever.

There is no fact of the matter there: it’s just a question of what you want to believe. There is a consensus tree, there is a majority tree, there is some kind of central tendency of the data, but whether that central tendency reflects historical branching or instead reflects ecology is another question.

Imagine you started out with three separate populations of things, and they just started exchanging genes with each other, and population A changed genes with B more frequently than they do with C. Eventually, then, most of the genes that you might make a tree of would show A and B as being sister with each other and imply that there was a common A-B ancestor, but there need never have been a common A-B ancestor. It is possible that you could get quite a robust tree of life, but it doesn’t necessarily mean that the nodes in that tree correspond to ancestors in the history of life.

Gitschier: Now with all the many genome sequences available, what does Doolittle posit about the relationship between Bacteria, Archaea, and eukaryotes?

Doolittle: I think there are two groups of prokaryotes: Bacteria and Archaea. They are not well defined, and there are many genes that are derived from lateral gene transfer from bacteria into archaea, somewhat fewer in the other direction. So to really say that “this bug is an archaeon” when the majority of its genes are actually bacterial, what you really mean is that you are privileging the ribosomal RNA as the definer. And that is what people do, so I will let them do that.

And then what people would believe, and I guess what I would believe, is that the eukaryotic transcriptional/translational machinery—the informational machinery in the eukaryotic cell—arose within the Archaea, more recently than the Bacteria and the Archaea diverged from each other. That would be the standard view.

But we think that a tremendous number of genes have been exchanged back and forth between bacteria and between bacteria and archaea, and also between bacteria and eukaryotes after eukaryotes arose from within the Archaea—so much transfer that it is really rather arbitrary to define these lineages by virtue of their transcriptional and translational machinery any more. Had Woese started looking at glycolysis enzymes, rather than ribosomal RNA, we might not even be talking this way.

There is a nice paper coming out in Nature by Bill Martin [Heidelberg] and others claiming that the various different archael groups arose by the importation of different bacterial genes at different times in the past; so in a sense, the Archaea are the universal acceptors.

One of the poster children of the Archaea has been the halobacteria; they live in the salt and they are pink. They are aerobes—they have to be because they are sitting there on the top of water—and we’ve known for a very long time that the cytochromes and other aerobic things are bacterial in nature in those archaea. Martin and his colleagues show that they have accepted a bunch of bacterial genes for aerobic capacity. The nice thing about that view is that it makes eukaryotes just another group of Archaea that accepted a bunch of bacterial genes.

The photo is from Christina Behme at a workshop in Halifax in July 2009 on "Questioning the Tree of Life." That's me having dinner on the first evening with Ford Doolittle (left), John Dupré (standing), and Andrew Roger (right)


judmarc said...

Lovely stuff, thank you.

peer said...

So, after all a reshuffling of preexisting genetic information is the basis of evolution...that is the creationists paradigm.

Jonathan Badger said...

"So to really say that 'this bug is an archaeon' when the majority of its genes are actually bacterial, what you really mean is that you are privileging the ribosomal RNA as the definer"

But it isn't just "ribosomal RNA" -- it is every other information-processing gene as well. *All* Archaea have eukaryotic-like transcription and translation machinery, and *all* bacteria don't. It's kind of irrelevant from the standpoint of evolution if metabolic genes get shuffled around if the information processing systems are intact. Evolution is by definition the transfer of information (genes) over time. If we started see strains of E. coli and the like with TATA-box transcription, then yes, I'd agree that the "tree of life" had fallen and like, but nobody has found such a bug despite thousands of genomes having been sequenced.

Larry Moran said...

*All* Archaea have eukaryotic-like transcription and translation machinery, and *all* bacteria don't.

Please post a list of all the ribosomal protein genes and which ones show more similarity between eukaryotes and archaebacteria. Please do the same with the aminoacyl synthetases. For those you might choose Doolittle's papers.

I'd also be interested in any references you have showing that your statement is true for all three of the RNA polymerase subunits.

Jonathan Badger said...

It's not about finding gene "similarity" (are we pheneticists now?) or even actual gene phylogenies supporting a archaeal-bacterial grouping (which can have a lot of explanations besides HGT for them such as unequal rates of evolution). The obvious eukaryotic nature of archaeal information processing isn't just something that can be argued on paper from sequence similarity or trees, but something that can be shown in the lab. As I mentioned, transcription in archaea uses TATA boxes, just like in eukaryotes and not sigma factors as in bacteria. And as for translation, it is pretty hard to ignore that archaea use methionine like eukaryotes and not formylmethionine like bacteria

Jmac said...

Here area some quotes from Doolittle interview:

"Doolittle: ....So I would say that Dalhousie deserves most of the credit for having proven the endosymbiont hypothesis, not that there aren’t multiple proofs."

I guess with age the memory tends to fail? Just few years ago in his famous paper I think in 2009 Doolittle was wondering where some genes came from in order for the endosymbiont hypothesis to be even close to a hypothesis...

This is my favorite part:


"I was very fond of Lynn. She was very much an anti-establishment person. Evidence wasn’t all that important to her. Everybody gives her great credit for the endosymbiont hypothesis. She also always maintained that spirochetes were the ancestor of what she called the “undulipodia,” the eukaryotic flagellar apparatus, for which there is no evidence, nor ever was any evidence, yet she held onto that belief her whole life."

All one has to do is to be Darwinist and everything else falls in to a place called blindness.

I haven't laughed like that in a long time. Lol

Anonymous said...


I know you're a professional imbecile, but this is too much. The endosymbiotic theory is not about the eukaryotic flagellar apparatus. There's a period in the paragraph you cite that indicates a change of subject.

Learn to read for comprehension you idiot. That would save you a lot of embarrassment.

Joe Felsenstein said...

What does Ford say about the tree among eukaryotes? Is the presence of a clear treelike signal among, say, the five species of the sparrow genus Zonotrichia the result of patterns of ecological similarity, or a hierarchy of descent?

When Doolittle and Bapteste, in 2007, declared that there was no tree, they quietly qualified it by saying "at least in prokaryotes". Creationist web sites featured their work and loudly acclaimed it, while deliberately ignoring the qualification. They also acclaimed the New Scientist's "Darwin Was Wrong" cover.

Has Ford, or for that matter our own Larry, expressed any disquiet about these misuses of statements about there being "no tree"? Or do they think that these statements are not misuses, that they are a correct assessment of the situation in eukaryotes?

Alex SL said...

That may all be very nice in prokaryotes, but you can't tell me that there is much horizontal gene transfer between, say, magnolias and roses, let alone between plants and animals. It does happen, rather rarely, but in my eyes the argument that this makes the tree of life not a tree is a few orders of magnitude less convincing than the hypothetical analogous argument that the existence of kidney transplants erases the family lineages of individual human beings.

Basically, what Joe Felsenstein wrote.

Tom Mueller said...

You have to admire Doolittle’s prescience!

This just in the news:

Deep-ocean microbe is closest living relative of complex cells

Genomic study of “Loki” supports a revisionist view of the origin of eukaryotes

Tom Mueller said...

Here is the original article:

and the Carl Zimmer article thereto pertaining:

SRM said...

Unless my jerk roti is what I hope it is, the evoluanoncence factory didn't work.

Seriously, Pest, do you pen all of your replies while you are stoned on mushrooms? I predict you would gain some small measure of respect around here, were this true.

Robert Byers said...

"its just what you want to believe" . Oh brother. Thats not science.
This tree cutting/pruning/uprooting is just a late realization that evolutionary connections are not real things ij nature. It doesn't work. DARWIN was wrong indeed.

All it is IS grouping traits. Yet doesn't allow for original boundaries of traits and doesn't allow for other mechanisms to bring out like traits.
Its all just a hunch and then grouping traits.
Its like with human skin colour.
A creationist must say human skin colour came from innate mechanisms to change populations very quickly. No selection but instant change upon migration to areas. Any like colour between segregated populations is just because of like mechanism bringing like change.
For example. All the white people groups only turned white after the language segregation and after migration to europe. Not from a original white tribe or tribe that moved into europe.
Evolutionists must instead say there is a tree of human colour. Each new population evolving a new colour from the pop it left. Sio they must connect peoples of the world by colour. Selection working upon a original colour. No innate changing.
Yet there is no such colour tree or any trees in biology. Its just lines of reassoning from received presumptions.

Rolf Aalberg said...

Not being a scientist I sometimes comment no Byers althoug he really doesn't deserve any attention at all.

But when he says things like Selection working upon a original colour. No innate changing. he demonstrate his ignorance. And the ignorance is coupled with ideas about sudden changes in any species according to his role as a YEC. As for instance his infamous claim that the thylacine was, not just a single wolf, but a population of wolves suddenly metmorphising into thylacines.

I just want to point out to Byers that there are variations within populations (the scientifc term is 'distribution of alleles within a population" or somthing like that. Not all alleles share the same likelihood of being preferred in the ongoing process of reproduction. Some alleles may be more succssful to the detriment of other alleles, making the population drift in a certain direction. Thus, an allele for light skin in an environment where lighter skin is advantageous with respect to reproductive success, the population as a whole will over time display a lighter skion colour.

Byers may tell us why that is impossible.

Larry Moran said...

The "Tree of Life" refers specifically to the attempt to unite all the major clades into a giant tree-like model that shows descent from a common ancestor. It is a tree that's supposed to represent species evolution. Recall that the Three Domain Hypothesis was promoted as a correct representation of the evolution of eukaryotes.

In that sense, there is no unique tree that accurately describes the evolution of eukarotes. Thus, it is correct to say that there is no tree of life while there still might be accurate large branches that really are tree-like.

We don't know exactly when the first eukaryote evolved but it seem likely that all eukaryotes descend from a common ancestor that lived only two billion years ago. There may be a strong consensus tree that unites all eukaryotes into a single tree-like model of species descent. However, this is not a tree of life—it only represent the last half of the history of life.

Also, keep in mind that so far there is no consensus eukarotic tree. The vast majority of eukaryotes are single-cell organisms and nobody has been anle to establish their relationships with each other. It's easy to get fooled if you are only thinking about animal trees or flowering plant trees. This is mostly the top 25% of the tree of eukarotes.

I did not object to the idea that there was no tree of life but I did object to the idea that Darwin was wrong.

Georgi Marinov said...

I don't like the word "revisionist" - it has been apparent for a while that eukaryotes came from within the TACK archaea, this was just the last piece of evidence needed to clinch it (and you can be sure Doolittle knew about it when he was interviewed).

Larry Moran said...

It had also been apparent for more than a decade that even more nuclear genes in eukaryotes come from within alpha-proteobacteria.

Tom Mueller said...

... there is no unique tree that accurately describes the evolution of eukarotes.

OK – I do not want to cross-post, so I will refer back to this post…

… while responding

About 82% of all ancient eukaryotic genes are more closely related to eubacterial genes than to archaebacterial genes. About 15% are closer to archaebacteria and 3% are ties.

Many, but probably not all, of the eubacterial genes come from the endosymbiotic event that led to mitochondria.

Your task, should you choose to accept it, is to draw a TREE that accurately reflects those facts.

I guess the part I am having difficulty with is the notion that a “unique” Tree cannot have “fusion points” our genealogy has fusion points, therefore we cannot construct a unique tree.

I like Simon Gunkel’s metaphor of Charles Darwin’s genealogy.

Heck, there are web-like structures in the biblical genealogies and of course Darwin children could trace back to their great grandparents in two different ways, which makes the Darwin household a really good illustration of how non-tree topologies work in biology.

To my way of thinking, the TOL as a historical sequence of events remains unique. What is being demanded remains a bridge too far. Is Josiah Wedgewood the paternal great-grandfather or the maternal great-grandfather to Darwin’s children?

I agree there is no unique “label” or description to "Josiah Wedgewood" from the great-grandchildren’s POV. That said, the genealogical tree as is, still remains unique.

So what am I missing?

Tom Mueller said...

oh dear - surely it cannot be true!

Is it possible that researchers have succumbed to the temptation of over-stating their case and would be guilty of hubris and hype?!

Georgi Marinov said...

Laurence A. MoranSaturday, May 09, 2015 9:54:00 AM
It had also been apparent for more than a decade that even more nuclear genes in eukaryotes come from within alpha-proteobacteria.

Where have I disputed that?

Tom Mueller said...


Larry has a great series of pertinent links right here:

I will repeat my original question therein:

...would recent hubris regarding the complexity of Pandora Virus be yet another example of self-promotion


is there merit to the suggestion that Pandoravirus presents cogent support for the suggestion of a fourth domain???

Georgi Marinov said...

The whole "How many domains?" argument is pointless. For the record, there isn't really merit to the suggestion that the giant viruses are a fourth domain, they seem to be just NCLDVs that have acquired a lot of genes

But whether there are three domains or two is just semantics. The argument that the Lokiarcheota are important for is was the ancestor of eukaryotes an archaeal lineage that branched from within the extant archaeal diversity (within the limits to which we can talk about branches in prokaryotes) or a lineage that branched earlier and then died off. It seems like the former is true. That's a clearly defined question thus it makes sense to pursue it aggressively.

The domains one is not so clearly defined because what the criteria for defining a domain are are not clear.

Tom Mueller said...

Hi Georgi

again and as always, thank you!

I remain in your debt

any chance you could set me straight on

I must be missing something

Georgi Marinov said...

Koonin has written a series of papers trying to quantify the tree-like and the net-like component in evolution:

And his conclusion is that there is a substantial tree-like trend even if HGT is rampant.

As far as what happened within eukaryotes, there is definitely a lot less HGT of prokaryote type there, but there is quite bit of EGT (endosymbiotic gene transfer), and disentangling the history of that will be very difficult, especially if it turns out, as have been suggested, that there were many cryptic endosymbionts in the history of some of the major lineages.

But I am not ready to accept that it will be impossible to resolve a tree for eukaryotes. Many of the problematic taxons are problematic because they are just not well characterized genomically yet. It does not guarantee success but often, when a whole genome, or at least a transcriptome becomes available, such lineages begin to resolve better, and once you have many genomes, the picture improves considerably. We'll see where we will be in 10 years.

Tom Mueller said...

Hi Georgi, I have spent a pleasant Saturday perusing the links you sent me.

I thank you!

Please tell me if I am getting this right.

Woese’s TOL was in fact a ribosomal gene TOL thereby begging the question of any presumptive extrapolation to the species level if indeed the term “species” can possess any coherency when extensive HGT or endosymbiosis (cryptic or not) is in fact prevalent.

Phylogenomics provides a great diversity of tree topologies when looking at other genes and gene families. Extrapolation to species becomes not only futile (because there are so many different and contradictory trees) but in fact meaningless at a fundamental and philosophical level. No single gene tree can mirror the evolution of the entire genome or organism. So the entire notion of TOL is essentially still-born.

As a result, some iconoclasts in the scientific community declared the TOL to be uprooted or dead.

I hope I am getting this all correct so far.

The fact remains that organisms (individual cells at a minimum) and their genomes are not in fact philosophical abstractions and these in fact do have a lineage i.e. an evolutionary history. The fact that gene TOLs may not always correspond well to species TOLs does not ipso facto render further inquiry organisms'/species' lineages moot or meaningless.

(At least that is how I understand Koonin’s defense of Woese)

Koonin proposes that a central trend or statistical model that could at least in principle be deduced by judicious, comprehensive, and simultaneous comparison of various tree topologies and this statistical model can in fact coalesce to a consensual (and genealogically accurate) vertical Tree of Life with horizontal connections here and there (more so at the base and less so at the apex).

To summarize, even though Woese’s original notion of the TOL may be dead; Long Live the Tree of Life!

I wonder what mark I would have obtained with this answer on Larry's exam if I concluded that Woese's original TOL may be vindicated (with modifications depending how the Loki data pans out).

Georgi Marinov said...

A major reason why Woese used rRNA was that rRNA was all that was available at his time. It's most abundant and as a result easiest to characterize.

However, that does not mean giving rRNA such a privileged place is not warranted - as it turns out rRNA is indeed a good phylogenetic marker because it is not a subject to HGT (there might be some, but I am not aware of a single case of HGT-mediated rRNA displacement). The reason it is not on its own sufficient to resolve the whole tree is that you need more data than just rRNA to do that. And the problem with using just rRNA is that the rest of the genome is subject to rampant HGT, but with a central trend and a (very small) core set of genes (with mostly information processing functions) that are rarely displaced.

The discovery of Loki refutes the topology of the Woese TOL, because Woese drew the tree as having two separate Archaeal and Eukaryote branches but now it turns out that the Eukaryotes are nested within Archaea.

The concept of TOL has been attacked because of the prevalence of HGT within eukaryotes, and that was a perfectly justifiable. The problem with that is sociological and had nothing to do with science - every time some evolutionary concept needs rethinking, that is used by creationists as "evidence" that evolution is bunk. So one has to be very careful when making statements of that sort because for the average person on the street the message that "those dumb scientists were wrong once again" is very easy to understand, while the minutiae of the actual scientific question are completely beyond reach.

Tom Mueller said...

Hi Georgi

Yet again I must thank you for your patience and your indulgence. I have learned a great deal today.

However, that does not mean giving rRNA such a privileged place is not warranted - as it turns out rRNA is indeed a good phylogenetic marker because it is not a subject to HGT (there might be some, but I am not aware of a single case of HGT-mediated rRNA displacement).

Understood! I was attempting to state the thesis of the iconoclasts as best I could

Larry Moran said...

I think 16S RNA and 23S RNA give different trees. Is this true?

Also, it may be impossible to detect HGT with 16S because that tree is usually the reference tree used to detect HGT in other genes. There's no reason to believe that ribosomal RNA genes are immune to HGT. The other problem is that there are often multiple copies of ribosomal RNA genes in prokaryotes and some of them differ significantly.

Georgi Marinov said...

Different trees for which groups?

Claudiu Bandea said...

It is relevant to reiterate here some of the points about the Tree of Life (TOL) from a previous post (

Simon Gunkel; Friday, May 08, 2015 3:25:00 PM:

”The TOL is a phylogenetic concept it rests on genealogy and genealogy alone.

Tom Mueller; Friday, May 08, 2015 9:39:00 AM:

“The conclusion becomes simultaneously trivial and inescapable: There is in fact only ONE TOL! Figuring it out becomes the problem.”

A few years ago, I made similar points: (

“The intent of the TOL, however, is to establish the line of descent among groups of organisms, or species, not necessarily the evolutionary relationships among their genes. Certainly, each of the millions of cellular and viral genes has an evolutionary history that can be revealed by a sequence-based phylogenic tree, but many of these gene-based trees do not represent a TOL that reflects the line of descent among the species. The problem, therefore, might not be with the TOL but with the reductionist approach of generating a TOL based exclusively on sequence-based phylogenetic analysis.”

So, if there are no reasonable arguments against this paradigm from the readers here at Sandawalk, and from the endless number of evolutionists who have attempted to ‘poison’ the TOL during the last few decades, then we can conclude that the TOL is growing just fine.

Claudiu Bandea said...

To continue my comment above, obviously, there is room for getting philosophical and metaphorical when talking about the TOL, but ultimately we have to represent the facts. So, I have no problem with Ford Doolittle statement that:

“And really the question is what do you mean by the “tree of life?” And clearly—this is sort of Eugene Koonin’s position, which he calls the “forest of life”—there is a consensus signal.”

As a matter of fact, I feel somewhat guilty that I might have started this trend on a few years ago (, when I wrote:

“Outlining the history of life requires broad ideas and scenarios. Metaphorically speaking, only by developing broad evolutionary scenarios, it is possible to see the forest for the trees. And, ultimately, the objective of this article is to draw a fundamental map of the forest generated by life on Earth - a forest that we ironically call the Tree of Life.”

Larry Moran said...

@Georgi Marinov

We are discussing the tree of life so I'm interested in knowing whether the 16S and 23S trees show the same relationships between eukaryotes and prokaryotes. Nobody disputes the fact that the higher you go in the tree, the more consistent the data using different genes.

Joe Felsenstein said...

So when someone is discussing whether or not there is a tree of all the species of primates, they are to be shown Ford Doolittle's articles, which tell them "there is no tree of life"?

Georgi Marinov said...

That can be easily checked. I was asking whether you said that because you know of particular examples

Tom Mueller said...

Correct me if I am wrong, but I thought that Eukaryote rRNA (if not Prokaryotes) was in fact considered "immune" to HGT for a variety of complicated reasons.

Tom Mueller said...

@ Joe

Is it possible that we are debating the phylogenomic equivalent to the “Ship of Thesus Paradox”?

That is how I understand the issues posed by “Tree of one percent” by Dagan and Martin.

If so, then I suggest we may be debating semantics along the lines of Lenin’s famous and less than subtly obscene Весь вопрос — кто кого опередит?,_whom%3F

The fact remains that we, can in principle, trace a genealogical history even if there exist alternate individual paths from ancestors to descendants.

As you point out, (I hope I am understanding you correctly) Species Trees can even then still remain coherent. One needs just simultaneously trace all the parallel paths to any species lineage in question and worry about “identity” or “labeling” issues later.

This strategy is possible if Koonin's thesis is correct.

Tom Mueller said...

@ Claudiu


Any chance you want to weigh in on "Pandora Virus" providing support for a "Fourth Domain"? ... a suggestion that was summarily shot down in the reference Georgi provided.

Claudiu Bandea said...

Tom Mueller; Sunday, May 10, 2015 9:12:00 AM:

“Any chance you want to weigh in on "Pandora Virus" providing support for a "Fourth Domain"?”

Well Tom, if you insist…. but I need to address it in at least 2 relatively long comments.

First, let’s put things into a historical/scientific perspective. Despite the fact that viruses are the most abundant life forms on Earth and the repertoire of their genes is greater than that of cellular organisms (not to mention their extraordinary medical and ecological significance as well as their critical role in shaping the evolution of cellular species and their genome), the conventional paradigms about their nature, origin and evolution are in their ‘infancy’, to put it mildly.

Indeed, in what might be one of the most enduring misconception in biology, ever since they were discovered more than a century ago, viruses have been conceptually misidentified with the virus particles and defined based on the physical, biochemical, and biological properties of these particles.

So , not surprisingly, Jean-Michel Claverie, one of the leading researchers in the field of giant viruses, who was intimately involved in the discovery and characterization of Mimiviruses, Pandoraviruses and Pithoviruses, recently asked “what if we have totally missed the true nature of (at least some) viruses?”

And, to the consternation of many scientists in virology and associated fields, he answered the question in a rather revealing way: identifying viruses with the virus particles, he wrote, might “be a case of ‘when the finger points to the stars, the fool looks at the finger’.”

A more detailed discussion as well as pertinent references can be found in my PubMed Commons notes associated with his papers on Pandoraviruses ( and Pithoviruses (

In summary, it is a very difficult task to address the origin and evolution of viral lineages when their nature might be on shaky scientific grounds; on the other hand, though, their origin and evolution might be critical in shedding light on their nature. So, what’s the solution?

Claudiu Bandea said...

Fortunately, there are only two broad ways of thinking about the evolution of viruses: (i) they have evolved from simple to complex by increasing the size of their functional genome, or (ii) from complex to simple by reducing the size of their genome.

The current prevalent hypothesis is that viruses originated from simple genetic elements before the origin of cells. According to this hypothesis, these mysterious ancestral viral elements evolved by acquiring new genetic material, including genes coding for components of translation machinery, into complex viruses such as poxviruses, chloroviruses, mimiviruses, pithoviruses and pandoraviruses, whose genome is several times larger than the genome of numerous cellular organisms, including many eukaryotic organisms. Notably, and relevant to our discussion here about the Tree of Life (TOL), in this evolutionary model, the line of descent-based phylogenetic tree of viruses is independent of that of cellular lineages (see Fig 4 in my paper:, so if this model is correct, we need to accept two line-of decent TOLs, one for viral organisms and one for cellular organisms.

On the contrary, my fusion model claims that viral lineages originated from parasitic or endosymbiotic cellular species that started their intracellular development by fusing with their host cells and evolved by reductive evolution, that is by losing genetic material, into a myriad of viruses with diverse complexity, genome size, and life cycles. This model is strongly supported by the fact numerous extant viruses, such as poxviruses, mimiviruses, pithoviruses and pandoraviruses, start their intracellular development by fusing with their host cells (see Figs 1 and 2). According to this evolutionary model, numerous viral lineages originated from parasitic cellular species throughout the history of life and, therefore, they form distinct branches on the line of descent-based TOL (see Fig 4, panel III).

Although, similar to their free-living relatives, the extant intracellular parasitic or endosymbiotic cellular organisms (and there are thousands of them) do acquire some new genetic material, there is overwhelming evidence that, overall, these lineages have evolved by reductive evolution. Basically, this is a fact (for a more detail discussion and relevant references see my paper and separate comment at:

So, in context of the clear evidence that all intracellular parasitic or symbiotic cellular lineages have evolved toward smaller genomes and lower complexity, it is critical to address the question: why would viral lineages evolve in any other way within the same intracellular environment? As noted in my comment, without addressing this essential question, the papers on the origin and evolution of viral lineages are incomplete.

That being said, although we might think of viruses phenetically (i.e. as it pertains to their biological type of organization) as a "Fourth Domain", from an evolutionary perspective (i.e. phylogenetically) according to my model, they do not represent a "Fourth Domain."

So, Tom, I do agree with the suggestion made by the reference you mentioned that Pandoraviruses and other complex viruses do not support the existence of a "Fourth Domain", but I think that the rationale used in that reference for reaching to this suggestion was flagrantly flawed.

Larry Moran said...

Joe Felsenstein asks,

So when someone is discussing whether or not there is a tree of all the species of primates, they are to be shown Ford Doolittle's articles, which tell them "there is no tree of life"?

No. That would be stupid.

Georgi Marinov said...

The other problem is that there are often multiple copies of ribosomal RNA genes in prokaryotes and some of them differ significantly.

I forgot to ask about that too - can you give me examples?

The one time I have had to seriously work with a prokaryote genome, it turned out it had two sets of rRNAs because whoever sequenced it, did not notice the culture was contaminated and the assembly was a mix of two completely unrelated bacteria. So we had to resequence it from our own culture and then the second set of rRNAs disappeared. I wonder how much of the cases like that in the databases are due to such trivial reasons given that the first one I got my hands on had that issue...