The Tree of Life Consolidates

Roland Piquepaille writes "The Tree of Life is an expression first used by Charles Darwin to describe the diversity of organisms on Earth and their evolutionary history. There are only two life forms, — eukaryotes, which gather their genetic material in a nucleus, and prokaryotes, such as bacteria, which have their genetic material floating freely in the cell. Until recently, eukaryotes, which include humans, were divided into five groups. But now, based on work by European researchers, the Tree of Life has lost a branch. After doing the largest ever genetic comparison of life forms they concluded that there are only four groups of eukaryotes."

Comments

Science is a moving target

[davidwr]

The more we know, the more we know that what we knew was wrong.

Or, as a coworker of mine used to say when we realized we didn't know what we were doing: "Everything you know is wrong."

Oops, my bad

[sm62704]

I should be more careful with that chainsaw. Poor tree, only 5 branches , I hope it survives...

First we lose pluto and now we lose this.

[Anonymous Coward]

What else will science rob from us before we decide enough is enough?

Not really a tree...

[myowntrueself]

While a tree-structure is algorithmically convenient and very enticing... the "tree of life" is not a tree.

Ie it is not a "directed, acyclic graph".

Unfortunately it has 'cycles'.

Blame retroviruses; they can take genetic material from one species and insert it into the genome of another thereby creating cross-branches.

As I recall, from my genetics days, baboon retroviruses are a great example of this. Again, IIRC, domestic cats and humans both contain fragments of baboon retroviruses.

Its possible that the "Cambrian explosion" is a sign of the appearance of retroviruses on the scene.

The thing is that it is significantly harder to reason about graphs; trees are so much easier to deal with.

So its very tempting to see things like this as trees and to 'simplify out' the nasty cross-branches.

(I've studied genetics, computer science, logic and discrete math)

Archaea

[virology-not for com]

Let's not forget that many scientists think there are three domains (Prokaryotes, Eukaryotes and Archaea). Archaea are very similar to Prokaryotes in that they don't have a nucleus, but they also share many features with Eukaryotes, including several key enzymes. Due to their similarity to the two other lineages, it is thought that Archaea may in fact be the grand daddy of all life. http://en.wikipedia.org/wiki/Archaea

Re:Archaea

[reverseengineer]

The level of organization being discussed in the paper treats a subdivision of the eukaryotes into "superkingdoms." (There's actually not a completely agreed upon term for this level.) This would put these groups a level below the three domains (Eucarya, Eubacteria, Archaebacteria) proposed by Carl Woese. There's a high-level differentiation between the superkingdoms involved based on organization of flagella, with a high-level split between unikonts (one flagellum) and bikonts (two, naturally). This is of course based on evolutionary ancestry- humans are unikonts, but don't have many cells with flagella.

The unikonts contain the amoebae lineages in one grouping, and the animal and fungi together in another. The bikonts contain the plants and algaes in one grouping, and also a handful of other groupings which take care of the rest of the eukaryotes, most of which are unicellular organisms of various sort. It is the "various sort" that's being ironed out with this paper- the authors argue that on the basis of a common genetic heritage, a couple of the leftover groupings can be consolidated.

Ironically, this move would actually reunite groupings that were fairly recently separated by the argument that no firm evidence of relation existed. Back when the "five Kingdoms" (Animalia, Plantae, Fungi, Monera, Protista) were considered the top level of organization, Protista existed as a sort of "junk drawer" for simple organisms which did not clearly fit in the other categories. Now it looks as though some of these organisms really are related.

Proof?

[Amorymeltzer]

Anyone who says "Evolution is taken as faith" or doesn't understand that the theory is based on the evidence, and that new evidence means changing the theory can look at this and shut up. A rather fundamental point was proposed to be rather fundamentally different based on new research and that's just fine. Whether it pans out or not, this is a beautiful example of the glory of science.

Re:Proof?

[MightyMartian]

Define "seen"? Because by your argument, electrons may or may not exist, Proto-Indo-European may or may not have existed and you may or may not have had great-great-grand-parents.

Evolution is confirmed not just by observing what goes on now, but by observing the fossil record, and just as importantly nowadays, by gathering molecular data. These two lines of evidence fit very well together into the so-called twin-nest hierarchy.

If you wish to wander down the road of epistemological nihilism, that's your affair, but be aware that everything, and I mean everything you think you know you can't actually know at all. Either you admit that inference is a legitimate means of gathering factual knowledge, or you render the whole show, including what you see, hear, touch, feel and taste irrelevant.

Taxonomy especially is inexact.

[verbalcontract]

When I was in 9th grade (I guess about 10 years ago!), there were five "kingdoms": bacteria, protista, fungi, plantae, and animalia. Three years later, there were six: archaea, monera, protista, fungi, plantae, and animalia.

Now there are branches? And four of them? On a tree? That's news to me. But it's all a matter of naming and grouping, so I guess you say potato, I say tomato.

At the risk of being called a Tao-rrorist...

[KublaiKhan]

In the pursuit of knowledge, every day something is added.
In the pursuit of understanding, every day something is removed.

Ummm....

[Otter]

The Tree of Life must be re-drawn, textbooks need to be changed, and the discovery may also have significant impact on the development of medicines.

This is a bit over the top. It's not like there's a single canonical "Tree of Life" that's going to have to be changed across the board; there's endless (mostly self-promoting) squabbling over what should be considered fundamental branches, to which this is yet another entry.

Frankly, if this were as important as they make out, it would be in Nature, not the if-it's-not-objectively-wrong-it's-in PLoS ONE.

"...the largest ever genetic comparison of higher life forms on the planet"? Maybe, I guess it depends what dimension you measure "largest" on.

Three levels of truth (maybe more...)

[gillbates]

Yes, being able to correct mistakes is the glory of science. But being right the first time is the glory of religion.

When religion doesn't get it right, people abandon it completely. When science doesn't get it right, they say, "well, that's just part of the process..."

Each particular method has its strengths and weaknesses:

• Religion reveals the truth of divine revelation. Which means that it is true by axiom, not proof. If the "revealed truth" isn't actually true, then it isn't of divine origin. Which does much to explain why religious institutions are very conservative when it comes to accepting new ideas.
• Mathematics is provably correct. That is, apart from an error in the proof, what is true today will always be true.
• Science is experimentally correct. That is, the hypotheses called true today may be shown false tomorrow with the discovery of additional data.

The key, I think, is not to confuse the various levels of truth. Those who take religion as if it were a scientifically-verifiable fact are just as confused as those who think scientific theorems are as reliable and trustworthy as the Gospel or mathematical proofs. There is a large difference between the three, and understanding the subtle limitations of each is just as important as understanding the ideas they espouse.

Re:Three levels of truth (maybe more...)

[NIckGorton]

When religion doesn't get it right, people abandon it completely. No they don't. They just reinterpret the primary tenets of the religion to suit their current desired conclusions. Religious works and religious beliefs are interpreted in the light of the present society and its prejudices. Rather than being taken at face value, they are used to justify what people want to believe. For example, there is no real prohibition against abortion in the Christian Bible. For another example, the selective interpretation of Leviticus as condemnation of homosexuality while ignoring the condemnation of poly-cotton blends and Red Lobster.

Well, unless you are someone who strictly interprets the OT: http://www.godhatesshrimp.com/

Re:Three levels of truth (maybe more...)

[zulater]

But many denominations do try to interpret in context and in light of the culture they were written in. Not to mention that the old law (old covenant/testament) was fulfilled after Christ's death and resurrection. That started the new covenant/testament which has no real limitations on what you eat. This isn't to say that there aren't valid teaching in the old testament but that the rules and regulations for ceremonial cleanliness don't apply anymore since there is no longer a need for sacrificing to cleanse sin. Abortion is murder in the eyes of the church because in most cases the church believes that life begins as conception or when the zygote attaches to the uterus. Many passages speak of God forming the child in the womb (Job 31:15, Psalm 139:15, Isaiah 44:2, Isaiah 49:5 etc). I agree that many do take the scriptures out of context and try to twist them into what they want to hear not what is actually meant. This is the folly of man trying to interpret scripture outside of other scripture.

Re:Three levels of truth (maybe more...)

[yali]

When religion doesn't get it right, people abandon it completely.

No they don't. They just reinterpret the primary tenets of the religion to suit their current desired conclusions.

A good demonstration of this is in the classic study When Prophecy Fails. A group of social psychologists studied a doomsday cult whose leader had predicted the end of the world. When the predicted date passed and the world didn't end, people did not leave the cult. Instead, they found reasons to explain it away (God was so impressed with their devotion that he put off the apocalypse on their behalf). The end result was that their beliefs were strengthened, not weakened, by disconfirmatory evidence.

(As a sidenote, the study was an important early test of Leon Festinger's theory of cognitive dissonance; Festinger had predicted the cult's response based on his theory.)

Re:Three levels of truth (maybe more...)

[JeanPaulBob]

For example, there is no real prohibition against abortion in the Christian Bible.

There is also no real prohibition against shooting people in the Christian Bible.

For another example, the selective interpretation of Leviticus as condemnation of homosexuality while ignoring the condemnation of poly-cotton blends and Red Lobster.

Far be it from me to interrupt your game of "Bash the Fundies", but...

The "condemnations" of homosexuality on the one hand and shrimp on the other are not the same, using entirely different words. (Just because the 400 year-old language in the KJV uses the word "abomination" in both passages, doesn't mean the Hebrew is the same.)

That raises the question, why do you make the peculiar assumption that every command in the OT law is of the same type, for the same kind of reason? Do you allow no distinction between ceremonial rules, and rules involving inherent moral/ethical concerns? Do you think that ancient Hebrews viewed dietary laws (prohibition of shrimp) and the command about mixed fabric as moral issues, in the same sense as murder, adultery, theft, and injustice? If so, why? If not, why base your arguments on absurd equivocation?

Re:Three levels of truth (maybe more...)

[MorePower]

That raises the question, why do you make the peculiar assumption that every command in the OT law is of the same type, for the same kind of reason?

Because the basic evangelical argument is that "morality" is based solely on "whatever God said, and humans dare not even try to ask why". If you allow for humans to have some capacity for independent moral awareness, then you would have what us heathen non-believers have been calling for all along, using our own sensibilities to decide what is and isn't acceptable.

I mean, how else do you condemn homosexuality or pre-marital sex? It's two consenting adults enjoying each other's bodies in mutually pleasing ways without harming others. But the evangelical crowd says "God said 'No', end of discussion."

But the circle of life...

[fahrbot-bot]

...stays the same:

• Born, Live, Read /., Die

Non-geeks, substitute "Read /." with "Move Out" and "Have Sex".

Re:So, what you are saying is.....

[MightyMartian]

There's a small part of most organisms' genomes that are made up of ERVs. These are insertions of retroviral DNA into our genomes. For the most part, these viral sequences are in neutral or junk genome stretches, so they don't have any influence on the organism. Unlike what the poster is saying, these don't make producing the tree more difficult, but in fact are extremely useful in fine-tuning the tree.

The odd-man out here are some prokaryotes, such as bacteria, where a sort of pseudo-sexual reproduction can take place by direct genome transfers. Still, this does not stop the classification of bacteria, but it does probably mean that the root of the tree of life, those earliest primitive self-replicators, probably swapped genes a helluva lot, so there may be no common ancestor per se, but rather a nest of common ancestors who swapped chunks of their DNA, RNA or whatever the earliest genetic molecules were.

Rather, you haven't been keeping up.

[IdahoEv]

When I was in 9th grade (I guess about 10 years ago!), there were five "kingdoms": bacteria, protista, fungi, plantae, and animalia.

What's happened is that better information has rapidly come to the fore as genetic analysis have been done during the last 15 years. The tree has been revised several times.

The five kingdom model was already known to be wrong 10 years ago, but that information hadn't propagated to gradeschool and highschool textbooks yet. If you'd studied biology in college, your information would be more up to date.

These days there are three superkingdoms: Bacteria, Archaea, and Eukarya. (Bacteria and Archaea were formerly grouped together as "monera" or "bacteria" before it was realized that genetically they are as distinct from each other as they are from Eukarya.) Eukarya is broken into a number of kingdoms, and that number has just changed from 5 to 4. Even the 5 they were last year weren't exactly same ones that you learned in school.

No!

[waveformwafflehouse]

4 supergroups? Wait, did Journey break up? Who stopped believing!

Re:Just hoopla over definitions

[Seiruu]

[quote]I never heard of "PLoS ONE"[/quote]

Then you have also never heard of Open Access, because then you would certainly know what PLoS ONE is. A shame you've never heard of it, because it is a very significant and rapidly growing movement within the scientific community. It puts the emphasis on opening up the access of scientific literature to everyone by switching from reader-pays to author-pays models. And with that said, it is very likely that scholars select PLoS ONE or other OA journals (peer reviewed of course) to show that they believe in the Open Access concept and let everyone with a digital connection have access to it.

Re:Just hoopla over definitions

[Anonymous Coward]

I never heard of "PLoS ONE", it claims to be a peer reviewed journal at least. If this was ground breaking I'd expect it to be published in Nature though.

I'm actually surprised you haven't heard of PLoS journals.

PLoS is an open-access publisher of science journals. Basically, the journals are free to access, and content is published under a creative-commons-type license.

PLoS journals are excellent, and rival the best journals in their content. There is no "general" science journal like Science or Nature, but there are topical journals like PLoS Biology, PLoS Medicine, etc. I'd argue that the content in these topic journals are comparable to Nature and Science publications.

PLoS One is a relatively experimental journal published by PLoS that attempts to push the open access model to its limits, by making the peer-review process completely open, where anyone is allowed to comment.

If anything, the fact that this article was published in a PLoS journal raised rather than lowered my expectations regarding its quality.

Re:Pretty good guess

[RockDoctor]

5 instead of 4? Well that's a pretty good guess given that the previous generation of scientists did not have a lot of the tools that the new kids on the block have.

The previous "5 kingdoms" model is hardly the result of guesswork. I've been working through a (now-outdated) reference tome on the model on-and-off for about 4 years now, and I'm barely half way through the book (It's Margulis & Schwartz, BTW, "Five Kingdoms: Illustrated Guide to the Phyla of Life on Earth (Paperback) " ISBN 0716730278).
Given that it's 10 years old now, I was actually expecting this to happen. In the time since I got the book (about 5 years) and started working my way through it, making notes, one of the 137 phyla which they describe has been found to be a grossly degenerated member of another phylum (it's an obscure parasite found normally only on the gills of cephalopod molluscs), another two have been merged (I can't even remember which ones they were. Protoctists of some sort.), and now someone has proposed a different way of slicing up the pie at the super-phylum level. I see that the unikont grouping still stands in this new analysis, which even I could figure out as a natural grouping.

Trust me (or do the legwork for yourself!), the 5 kingdoms model was not guesswork. It might not be the correct model, but it's based on a lot of evidence.
(BTW, sitting in my rucksack at this very moment I've got a reprint of one of Margulis' 1995 papers setting out some of the grounds for the 5 kingdoms model. It's my "light reading" on the bus to work, as a change from doing a correspondence course in Java. Next to it is a reference to the geological structure of the South Atlantic, which may be my work place in a couple of years. Lifelong training is a requirement, not an option.)