Not exactly the Garden of Eden

March 24, 2010 2:11 PM

"People are going to be what we say 'gobsmacked' by this news," said Terry Brown. New human ancestor.
posted by archivist (58 comments total)

That landmass might have been a peaceable kingdom of competing "hominin" species. Or it could have been the site of genocide, with the Neanderthals and the just-discovered species dying out to the last man and woman.

I wonder which of those possibilities is more likely.
posted by Joe Beese at 2:14 PM on March 24, 2010

I'M ABOUT TO LOSE MY MIND IN HERE.
posted by Astro Zombie at 2:17 PM on March 24, 2010 [5 favorites]

I'm having trouble parsing the comparisons in the quote about the mitochrondrial DNA:

Modern humans differ from Neanderthals by an average of 202 "letters" out of about 16,500 in the complete mtDNA strand. The Siberian bone's mtDNA differs by 385 letters. Chimpanzees and modern humans differ by 1,462 mtDNA letters, on average.

Am I right in thinking that "The Siberian bone's mtDNA differs by 385 letters" means "The Siberian bone's mtDNA differs from modern humans by 385 letters" rather than "differs from Neanderthals"?
posted by Elsa at 2:25 PM on March 24, 2010 [1 favorite]

Have I got this right? They're only talking about mitochondrial DNA. So although this is a different matrilineal lineage, they might well have been indistinguishable from our mainstream ancestors or from Neanderthals both phenotypically and in terms of 'ordinary' DNA?
posted by Phanx at 2:28 PM on March 24, 2010 [1 favorite]

Elsa: The other two comparisons both involve humans, so I don't know. Of course, if so, that contradicts the earlier statement that this species is closer to human than Neanderthals were. Either way, I hope they don't intend to close anything with that DNA.
posted by The Winsome Parker Lewis at 2:29 PM on March 24, 2010

*if so = if the author's comparing the Siberian bone's mtDNA with that of humans
posted by The Winsome Parker Lewis at 2:30 PM on March 24, 2010

I don't know, Phanx. Aren't all modern humans supposed to have similar mtDNA? The assumption seems to be that variations in the mitochondria are an indicator that this species was separate from our line for at least a million years.

And boy, our ancestors must have been quite the killers! They polished off two competing hominid species and most of the large animals and birds in the entire world, and all it with stone tools. Just imagine if they'd had guns...
posted by Kevin Street at 2:36 PM on March 24, 2010

Man, no wonder Europeans are so much better at genocide than everyone else. They got an early start.

NOT ANTI-EUROPEANIST! Some of my best friends are white! Plus my family!
posted by ibmcginty at 2:37 PM on March 24, 2010 [3 favorites]

Obviously they're either elves or dwarves. Hard to tell from just a single finger bone, but we've already found the hobbits so that rules them out.
posted by XMLicious at 2:48 PM on March 24, 2010

oh sweet yeti party
posted by Greg Nog at 2:48 PM on March 24, 2010

I'm going to go with Atlanteans.
posted by Astro Zombie at 2:50 PM on March 24, 2010

There's something ironic in the idea that we wiped out our cousins a long time ago, then invented a whole bunch of spirit folk and little people legends because we felt lonely.
posted by Kevin Street at 2:53 PM on March 24, 2010 [24 favorites]

The best thing about the C64, I tend to think, was that when you turned it on, it was ready for you to write programs immediately.
posted by Wolfdog at 2:53 PM on March 24, 2010 [35 favorites]

oops, multiple tabs open, but that was probably a more amusing comment than the feigned mix-up about Rush producer Terry Brown that I was contemplating posting in this one, so it's probably all to the good.
posted by Wolfdog at 2:58 PM on March 24, 2010

Obviously they're either elves or dwarves.

It's the human g-nome project.
posted by It's Raining Florence Henderson at 2:59 PM on March 24, 2010 [22 favorites]

Cylon?
posted by sentient at 3:00 PM on March 24, 2010 [4 favorites]

Two observations: I'm not up on the most recent research, but I seem to recall that there's a good deal of uncertainty about the molecular clock hypothesis, i.e, that changes in mtDNA happen at a predictable rate and thus can be used to infer evolutionary histories. Second, as phanx notes, just because mtDNA differs substantially does not guarantee that the specimen comes from a distinct species.
posted by docgonzo at 3:04 PM on March 24, 2010

Why is Science publishing news stories on this while holding back the research paper for another couple of days?
posted by grouse at 3:06 PM on March 24, 2010 [1 favorite]

How are they ancestors if the line died out? Wouldn't that make them at most extinct relatives?
posted by longsleeves at 3:07 PM on March 24, 2010

Whoever this was, they were apparently crunchy and good with catsup.
posted by Danf at 3:08 PM on March 24, 2010 [1 favorite]

Just imagine if they'd had guns...

We don't have to, we're living in it.
posted by strangely stunted trees at 3:08 PM on March 24, 2010 [3 favorites]

Obviously they're either elves or dwarves.
----
It's the human g-nome project.

That made me snort a bit more loudly than I'm comfortable with.
posted by FatherDagon at 3:10 PM on March 24, 2010 [6 favorites]

We get this notch, right? Not the Neanderthals?



THERE CAN BE ONLY ONE......hominid species.
posted by Atreides at 3:11 PM on March 24, 2010 [1 favorite]

I wonder if we're going to reach a point where it will have been considered short-sighted to think of only one or two related precursor species, rather than several divergent lineages, most dying out.
posted by Durn Bronzefist at 3:13 PM on March 24, 2010 [2 favorites]

> And boy, our ancestors must have been quite the killers! They polished off two competing hominid species and most of the large animals and birds in the entire world ...

Not to mention the dinosaurs.
posted by brokkr at 3:14 PM on March 24, 2010 [1 favorite]

posted by Kevin Street And boy, our ancestors must have been quite the killers! They polished off two competing hominid species and most of the large animals and birds in the entire world, and all it with stone tools. Just imagine if they'd had guns...

Just imagine if they'd had a debate about health care reform.
posted by mattdidthat at 3:23 PM on March 24, 2010 [3 favorites]

grouse,

The paper is in Nature. There's a link at the end of this summary. (I think the summary is freely available; the research paper requires a subscription.)
posted by lukemeister at 3:29 PM on March 24, 2010

This is a better write up of the story IMO.

They got a big surprise: Although Neandertals differ from modern humans at an average of 202 nucleotide positions in the mitochondrial genome, the Denisova hominin differed at an average of 385 positions from modern humans and 376 from Neandertals, the team reports online today in Nature. When mtDNA from chimpanzees and bonobos was added to the mix, the researchers were able to estimate that the new hominin had shared a common ancestor with Neandertals and modern humans about 1 million years ago.

I think the human ancestor part is a bit of a stretch. A completely amazing discovery!
posted by batou_ at 3:29 PM on March 24, 2010 [1 favorite]

Oops, thanks, lukemeister.
posted by grouse at 3:30 PM on March 24, 2010

Maybe it's just 'cause I'm on a college campus but I can see the article right here.
posted by pwb503 at 3:36 PM on March 24, 2010

I miss them.
posted by darkstar at 3:46 PM on March 24, 2010 [1 favorite]

Ow!
My gob!
posted by bstreep at 4:10 PM on March 24, 2010

No pictures? Didn't happen.
posted by Cranberry at 4:26 PM on March 24, 2010

I wonder about the "genocide" hypothesis. It's often said that the arrival of humans in North American caused the extinction of megafauna such as mastodons and mammoths, but it's much more likely that this animals died because of changing climatic conditions and disappearing range.

Wouldn't intermarriage or even climate-related extinction also explain the end of this line?
posted by KokuRyu at 4:33 PM on March 24, 2010

Wolfdog, apologize for nothing: that dadaist non sequitur just made my day. I wish I were joking. Thanks for that.
posted by joe lisboa at 4:40 PM on March 24, 2010

Willerslev emphasizes that, on its own, the mtDNA evidence does not verify that the Siberian find represents a new species because mtDNA is inherited only from the mother. It is possible that some modern humans or Neanderthals living in Siberia 40,000 years ago had unusual mtDNA, which may have come from earlier interbreeding among H. erectus, Neanderthals, archaic modern humans or another, unknown species of Homo. Only probes of the nuclear DNA will properly define the position of the Siberian relative in the human family tree.

Do we have mitochondrial DNA from earlier humans? Given that there's only one bone and we don't know what they looked like, could this be an already known kind of human? I guess they would have ruled that out somehow, but it's interesting anyway in what it tells us about relatively recent human history: us being alone as the only kind of human is the exception, and not the norm.



Man, no wonder Europeans are so much better at genocide than everyone else. They got an early start.

NOT ANTI-EUROPEANIST! Some of my best friends are white! Plus my family!

I know this was total hamburger, but the bone was found in Asia. So, um.
posted by Sova at 5:16 PM on March 24, 2010

I actually came across this story as I was poking about on Fox News. At the time there were only a few comments, the bulk of which were along the lines of "Huh, science on Fox?" similar to my own reaction. (I've been surprised by some content on fox news recently until I noticed these stories are usually reproduced from AP).
posted by TwoWordReview at 5:21 PM on March 24, 2010

Actually, it doesn't sound like this was actually any sort of 'ancestor', just another branch of the hominid tree.
posted by delmoi at 5:23 PM on March 24, 2010 [2 favorites]

Sara: You're an animal!

Manny: No, worse! Human. Human!

~Runaway Train
posted by bwg at 6:04 PM on March 24, 2010

Just got back from a bar room chat about this paper; you beat me to a post. Before this study, we knew about three major hominin migrations out of Africa - Homo erectus around 1.8 million years ago, the ancestors of the lineage that eventually gave rise to Neanderthals around 500,000 years ago, and then anatomically modern humans around 50,000 years ago.

The coalescence dating of this sequence shows that it shares a most recent common ancestor with Homo sapiens and Neanderthals around 1 million year ago. For the identity of this finger's owner, that puts H. erectus pretty much out of the running: 1 million years ago, the ancestors of H. sapiens and H. neandertalensis were still in Africa, while H. erectus was busy populating the Old World. The divergence between our lineage and that of the finger-owner must have taken place in Africa, meaning that this lineage's migration out of Africa adds to the three we already knew about.

We have good sampling of anatomically modern human and Neanderthal mitochondrial DNA, so it's unlikely it's just a funky member of either of those groups. In the analysis they included an anatomically modern human Cro-Magnon from 28,000 years ago (Late Pleistocene). It grouped with the other humans, not surprisingly.

So, maybe this is Homo heidelbergensis, the antecedent to Neandertals and H. sapiens (probably). Maybe it's another hominin species-- I propose Homo pinkifingeronliensis. Having a bunch of hominins running around Eurasia is not all that surprising, really. The human evolutionary tree has been more like an evolutionary bush up until very recently, when it was pruned to just one branch.

What's beyond awesome is that they thought they were gonna sequence some DNA out of a boring Neanderthal bone, and ended up with the entire mitochondrial genome of one of our cousins, 1 million years removed. I can barely get DNA out of primate poo that's a few years old. The future is now.
posted by bergeycm at 6:19 PM on March 24, 2010 [20 favorites]

It's often said that the arrival of humans in North American caused the extinction of megafauna such as mastodons and mammoths, but it's much more likely that this animals died because of changing climatic conditions and disappearing range.

Really? So take the American mastodon, for instance - the species survives for almost 4 million years, through maybe a dozen or so glacial/interglacial cycles of equal intensity to the Holocene, and then they just happen to go extinct within 5000 years or so of humans populating the continent, but not because people overhunted them?

Yeah, I guess it's possible, but I don't see how you can make a case for it being much more likely, especially given our species' proven track record at driving megafauna to extinction in the historic era.
posted by strangely stunted trees at 7:06 PM on March 24, 2010 [3 favorites]

In case anyone missed these (I did since I incorrectly assumed I'd to have library access)...

This phylogenetic tree for the mtDNA is kind of helpful. It helps give an idea of what we're talking about here, and the justification for calling the Denisova a different species vs. an ancestor, etc.

And yeah, it doesn't look like an ancestor so much as another species/hominid with which Neanderthal and Humans share a common ancestor. (Excepting the caveats about nucleotide mutations and the corresponding evolutionary paths, etc.)

Here's a chart showing the pairwise nucleotide differences. (What's with the crazy bimodal distribution for human-human?)
posted by sentient at 7:06 PM on March 24, 2010 [3 favorites]

Ok, so there's another line of proto-humans who came out of Africa, had their own breeding patterns, made it to Siberia, and then died out.

I'd be surprised if there weren't more proto-human/human lines/lineages out there that died out.

Life is a sexually transmitted disease - you hear stories about people a couple/few generations back who have thousands of ancestors. Do you ever hear about the kids of the people of that generation who never had kids?

It's interesting, though, that the human genome is so tight that anyone from any culture is still able to produce viable offspring with one another. Although I have to speculate that it may be HLA discrrepancies that might cause early miscarriages in couples who are thought to be "infertile.")
posted by porpoise at 7:12 PM on March 24, 2010 [1 favorite]

So, maybe this is Homo heidelbergensis, the antecedent to Neandertals and H. sapiens (probably). Maybe it's another hominin species-- I propose Homo pinkifingeronliensis. Having a bunch of hominins running around Eurasia is not all that surprising, really. The human evolutionary tree has been more like an evolutionary bush up until very recently, when it was pruned to just one branch.

So how many different kinds of human do you think there might have been at any one time? 4? 5? More? As little as 40k to 50k years ago?

I find that amazing.
posted by Sova at 9:46 PM on March 24, 2010

So how many different kinds of human do you think there might have been at any one time? 4? 5? More? As little as 40k to 50k years ago? I find that amazing.

I have thought about this often. Imagine what the world would be like now if there were 2 or 3 hominid species around. It would be an episode of Star Trek or The Twilight Zone.
posted by neuron at 10:12 PM on March 24, 2010 [2 favorites]

John Hawks, as usual, is good on this. He concludes it is far to early to tell whether this is a new species.
posted by afu at 10:19 PM on March 24, 2010 [1 favorite]

So how many different kinds of human do you think there might have been at any one time? 4? 5? More? As little as 40k to 50k years ago?

Yeah, most definitely. We have fossil evidence for the presence of modern humans, Neanderthals, Homo floresiensis, and now this finger-owner around 40,000 years ago. Neanderthal fossils were found less than 100 km from this unknown hominin. Taking into account the paucity of the fossil record, there are probably more, especially those living in tropical environments where preservation is poor.

This isn't unprecedented. 1.5 million years ago, there'd be at least four good hominin species we know from the fossil record. Paranthropus robustus, P. boisei, Homo erectus, and Homo habilis. (There may be even more legit species, like H. ergaster.)

Long (6 million-year-old) story short: we're not special, just lucky.
posted by bergeycm at 10:35 PM on March 24, 2010 [4 favorites]

...then invented a whole bunch of spirit folk and little people legends because we felt lonely.

Spirit folk and little people being romantic and pleasant are 20th century conceits. Before then, they were terrifying monsters. Which is probably why there are no more little men and dwarves and trolls and such... and probably why the Uncanny Valley works the way it does. It looks human, it acts human, but it isn't human, it's awful and scary, KILL IT!

On the other end, I seriously doubt THEY were interested in having us over for the Book Club - genocidal paranoia is part of the evolution game.
posted by Slap*Happy at 11:22 PM on March 24, 2010 [4 favorites]

I was just about to link to the John Hawks article, which I saw on Carl Zimmer's site. Here's his article on the topic.
posted by delmoi at 3:47 AM on March 25, 2010

Yes, yes, but how did they taste?
posted by Pollomacho at 4:44 AM on March 25, 2010

"Maybe it is overly simplistic to think of particular migrations out of Africa," said Svante Paabo, the other leader of the German team. "There might have been a more or less continuous flow of migration."

Makes good sense to a layperson. Migrations of populations in the Americas didn't happen in single event spikes - they were gradual and constantly shifting.

Man, no wonder Europeans are so much better at genocide than everyone else. They got an early start.

By no means have European peoples conquered the market on genocide. I've been learning quite a bit about the Maya recently, and they enjoyed nothing more than smashing rival tribes to tiny little bits.
posted by Miko at 5:07 AM on March 25, 2010

Given the infinitesimally small number of animals whose bones get fossilized and the unlikelihood of someone ever finding them, it's amazing that we know anything at all about our ancestors.

And it seems highly presumptuous to assume that we have even a small fraction of the complete picture of human origins.

This is a fascinating find.
posted by rbellon at 5:39 AM on March 25, 2010

So how many different kinds of human do you think there might have been at any one time? 4? 5? More? As little as 40k to 50k years ago?

There are ubiquitous stories of black dwarves in Germany and Scandinavia and the British Isles. And stories of giants in Scandinavia. And in the Torah, giants and Nephilim. In Arabia, of djinn (some of whom are large and black) and efreets. Not to mention the stories of human-baby stealing pygmies in Indonesia.

I've long thought these legends must be rooted in some truth, and that truth is cultural memories of our meetings with other species of Hominina.
posted by orthogonality at 6:29 AM on March 25, 2010

These "Forerunners", did they love bees?
posted by Uther Bentrazor at 8:04 AM on March 25, 2010

I don't want to be a spoil-sport, but the meaning and application of mDNA techniques is hotly debated among anthropologists. There are folks that don't think it means what others think it means at all, and each group tends to gloss over the arguments of the other when presenting news like this.

So, take news like this with a grain of salt.
posted by clvrmnky at 8:09 AM on March 25, 2010

I don't want to be a spoil-sport, but the meaning and application of mDNA techniques is hotly debated among anthropologists.

So, what is the other side of the argument? Links?
posted by grouse at 8:18 AM on March 25, 2010

I'll try and explain some of the rationale for the controversy.

Background: mtDNA doesn't undergo recombination like the rest of your DNA. It only appears in the mitochondria of cells. Your mtDNA forms a single chromosome, and all the copies in all the cells in your body are theoretically identical. You always get your mtDNA from your mother -- it's in the mitochondria in the cellular fluid in the egg from your mom.

So, if everything went perfectly, you, your mother, your mother's mother, and so on would all have identical mtDNA. But if there are mutations when more copies of the mtDNA are synthesized, and if these mutated copies end up in an egg, then they'll be passed on to the next generation.

So you can see how the changes might begin to accumulate, and that comparing mtDNA changes is simpler than nuclear DNA changes because nuclear DNA (i.e., which versions of chromosomes you have) is expected to vary from one generation to the next, and mtDNA really isn't.

If you assume that there'll be some mtDNA mutations in each embryo anyone produces, then you can see how closer relatives would have more similar mutations. Let's say that a women has mtDNA with mutation A. She has two daughters. Both will probably have mutation A. But each daughter might inherit a new and distinct mutation. So Daughter1 might have mutations A and B, and Daughter2 might have mutations A and C. And the same thing happens with each generation of children. The result is that all the great great great (etc.) grandkids will have mutation A. And some other lineage beginning with a totally different woman with a totally different mutation (say, mutation Q) would consist of great great great (etc.) grandkids with mutation Q.

The upshot is that you can find clusters of mutations that are more similar and that are less similar, and these help you sort people (or Neanderthals, etc.) into different lineages. The more divergent the sets of mutations, the more distant the relatives. Humans and Neanderthals have really different mutations. Cousins have very similar mutations. People of similar ethnicities and ancestry have similar sets of mutations. (Which is the basis for projects like National Geographic's Genographic Project.)

So based on all of this, you can see where some problems might emerge. Here are a few:
- You and I can have similar mutations by chance. This is all about probability. Probability is pretty reliable, but clearly it's not 100%. (heh).
- Time is extremely relevant too. Assumptions and generalizations are made when it comes to the rate of mutation (e.g., the number of mutations per generation), which affects how we decide the simplest way to get from one set of mutations to another, and thus the decision about who's an ancestor and who's a descendant.
- Overall, determining the groups of related people is based on the most parsimonious "tree" (like the one here.) So we assume the smallest changes from one generation to another, and create a tree based on that. Here's an example of this:

If you have six different samples of mtDNA, and each mtDNA sample has one of the six sets of mutations listed here: { {B}, {AB}, {AC}, {ABN}, {ABCD}, {ABCN} }, then it seems simplest to assume that the lineage looks like:

..............B
........... /...\
.........AB...AC
....... /
......ABC
...../......\
ABCD...ABCN

But, really, the lineage could equally simply look like this:

..............B
........... /...\
.........AB...AC
................ |
...............ABC
............../......\
.........ABCD...ABCN

Or it could look like this:

.....................B
............... /........\
...........AB.............AC
........../.....................\
......ABC.....................ABC
...../......\................./.......\
..ABCD...ABCN......ABCD....ABCN

Or maybe like this:

........... B
........... |
...........AB
....... / ..... \
....ABC......AC
.....|............|
ABCD.......ABCN

Which is actually correct? Probability says the simplest. But really it depends on whether the simplest lineage is actual lineage, and it's also based on how old we decide that a sample with n mutations (or mutation reversals) has to be, as well as where geographically the ancestor had to be in order to fit in the tree correctly.

Also, like others have mentioned, we assume that mtDNA divergence correlates with physical or overall genetic/phenotypic difference. (It makes sense that it would, but, hey, it's all about probability.)

So these are some basic problems that arise with mtDNA-based determination of relatedness and ancestry.

Apologies if some or all of this is old news.
posted by sentient at 7:07 PM on March 25, 2010 [4 favorites]

So, what is the other side of the argument? Links?

Look at the John Hawks article I posted above.
posted by afu at 7:20 PM on March 25, 2010