August 13, 2012

Human-chimp divergence date pushed back in time

I recently wrote:
Actually, we are beginning to get a hold of the mutation rate -thanks to the ability to sequence full genomes- but we have absolutely no clue when human-chimp divergence actually happened, at least not within a few million years.)
The context of that post was to highlight the fact that many aspects of recent human origins research, such as estimates of effective population sizes or mutation rates were inextricably linked to calibration with the date of divergence between humans and chimps. But, that date was obscure: paleontologists didn't always seem to agree on what represented a member of the lineage leading to us or a common human-chimp ancestor; and there were circular arguments a-plenty, where genetic mutation rates calibrated on a particular human-chimp divergence were then used again to "prove" that divergence rate.

Well, at least now, thanks to a new paper by Langergraber et al., we have a better idea. It's clear now that humans and chimps diverged from each other a few million years earlier than previously thought.

And, of course there are implications on every event where calibration based on human-chimp divergence was at play. That has major implications for Neandertal-human divergence, as well as for human-human divergence estimates.

Together with the publication of the two conflicting papers on Neandertal admixture (or not), it appears that August 2012 has proven to be anything but a time for a little quiet vacation.

From Science's coverage of the new work:

Now, after a decade of analyzing the patterns of reproduction in chimpanzees and gorillas in Africa, Vigilant and former postdoc Kevin Langergraber—now a primatologist at Boston University—say they have the data they need. Working with almost 20 collaborators, the duo gathered data recording the age of mothers and fathers at the birth of 226 offspring in eight different chimpanzee communities in the wild, and 105 offspring of mountain gorillas from two different research sites in Africa—and they verified those relationships with DNA paternity tests on coprolites gathered in the field. As they report today in the Proceedings of the National Academy of Sciences, chimpanzee mothers ranged in age from 11.7 to 45.4 years at the birth of their offspring. The average age of reproduction was 25 years for females and 24 years for males, giving them an average generation time of about 25 years. Gorilla females ranged in age from 7.3 to 38 years when they gave birth, and the average generation time for both sexes was about 19.3 years. 
To get the mutation rates, the team divided the number of mutations between parents and their offspring (collected by analyzing DNA from coprolites sent to the lab in Leipzig) by the newly calculated generation times. The researchers got a high and low mutation rate for each species per year—rates that were slower than previously estimated using fossils to calibrate the molecular clock. When they applied the new rates to the history of all three species, they calculated that humans and chimps split earlier than expected—at least 7 million to 8 million years ago and possibly as early as 13 million years ago. They estimate the split between gorillas and the lineage leading to humans and chimpanzees to 8 million to 19 million years ago. Those dates have such wide ranges, Vigilant explains, because they assume the mutation rates seen today have been constant over time in all three lineages. So a key remaining question is whether mutation rates were faster in the past.

Incidentally, this might solve Chris Stringer's doubts about the antiquity of the Atapuerca hominins, because it is now possible to reconcile their 600,000-year old age with a genetic date of modern human-Neandertal divergence of 400-800 thousand years.

Any further comments on the paper will be posted in this space after I read it.

UPDATE (Aug 15):

This is a very exciting paper that finally addresses the autosomal mutation rate controversy.

Previously, this rate was assessed by picking a split time for the chimp-human pair, and counting sequence divergence, dividing it by this time. But, this method led to circular reasoning: paleontologists looked at the (very fragmentary and controversial) early fossil record and assigned some arbitrary date to the split.

Then, when new fossils turned up that looked like they belonged to the human lineage but preceded the supposed split, their inclusion in the lineage leading to humans was rejected, on account of them being too old, pre-dating the supposed split!

A different approach is to use the directly observed mutation rate in families. This has been made possible due to the fall in the cost of sequencing. Subsequently, one divides sequence divergence with the mutation rate to obtain generations. But, how does one obtain years, instead of generations? A measure of generation length is needed.

And, this is the exciting and remarkable part: members of the team behind this paper didn't just "guess" the generation length in primates but actually observed primates in the wild, assigning hard parental ages where they could, or intervals where they could not. This was a decades-long process that gave us the best estimates of generation lengths to date.

Armed with this knowledge (directly observed generation lengths + directly observed mutation rates), they were able to go back in time and estimate the human-chimp divergence. Their answer?

As I have said before: this has implications. The authors do not touch on recent human history, except to point out that the Neandertal-human split harmonizes better with paleoanthropology:

analyses of the Neanderthal genome indicated a population split between present-day humans and Neanderthals at 270–440 ka (40). This date appears to conflict with fossil evidence tracing the emergence of Neanderthal morphological characters over the course of the Middle Pleistocene in Europe (41). The earliest evidence for Neanderthal traits was proposed to date to 600 +/- 66 ka at the Sima de Los Huesos (Atapuerca, Spain) (42), thus predating the genetically estimated population divergence times, but this date has been disputed on the basis of both the apparent conflict with the genetic data and on stratigraphic grounds (43). However, even if the early dates for Sima are disregarded, it is clear that fossils from oxygen isotope stage 11 (around 400 ka), such as the Swanscombe cranium, already show clear Neanderthal traits (44). Using the new human– chimpanzee split estimate and assuming generation times between 25 and 29 y would push back the human/Neanderthal split to 423,000–781,000 y, resolving this apparent conflict. 

The authors also propose that:
Whereas the earliest fossil universally accepted to belong to the lineage leading to present-day humans rather than to chimpanzees, Australopithecus anamensis, is 4.2 Ma (31) and thus reconcilable with a molecularly inferred human–chimpanzee split time as recent as 5 Ma, the attribution of late Miocene (5–7 Ma) fossils to the hominin lineage has posed a problem. Our estimates make it possible to reconcile attribution of fossils such as Ardipithecus kaddaba (5.2–5.8 Ma) (32), Orrorin tugenensis (6 Ma) (9), and Sahelanthropus tchadensis (6–7 Ma) (10) to the hominin lineage with speciation times inferred from genetic evidence (Fig. 1). However, our estimates cannot address the controversy of whether specimens such as these truly belong to the lineage leading to present-day humans or to other, closely related lineages (11). 
And, there are calibration points in more recent times that the authors do not consider, but will also support the new mutation rate/speciation time. For example, Gravel et al. obtained a 51ky Out-of-Africa, and sub-30ky dates for the European-Asian split using a mutation rate of 2.38x10^-8. Using the lower mutation rate will bring the European-Asian split up to the 40 thousands (consistent with the first appearance of AMH in both Europe and East Asia), and the OoA event to 100ky, consistent with the appearance of the Nubian Complex in Arabia and the Skhul/Qafzeh hominins in the Levant.

The new mutation rate will also affect other dates. Li and Durbin estimated a 100-120ky Yoruba-Eurasian divergence, but only by using a 2.5x10^-8 rate and explicitly rejecting the slower direct rate. We know for a fact that the major part of Yoruba ancestry is shared with Eurasians in the post-70ky time frame (Y-haplogroup E is part of clade CT, and most Yoruba mtDNA belongs to haplogroup L3, of which Eurasian N and M are branches). In order for a 100-120ky (or 200ky+ with the direct rate) divergence to be believable, only one possibility exists: that Yoruba trace their ancestry to a combination of post-70ky E/L3-bearing people who were quite close to Eurasians but also of a group of Palaeoafricans that split off from modern humans well before 200ky. And, I won't even go to the implications about African hunter-gatherers that are divergent from Eurasians by an even greater amount.

Enjoy the calm before the storm: if people start using the slow mutation rate (as they now must), a lot of the old certainties about our origins will be toppled.

PNAS doi: 10.1073/pnas.1211740109

Generation times in wild chimpanzees and gorillas suggest earlier divergence times in great ape and human evolution

Kevin E. Langergraber et al.

Fossils and molecular data are two independent sources of information that should in principle provide consistent inferences of when evolutionary lineages diverged. Here we use an alternative approach to genetic inference of species split times in recent human and ape evolution that is independent of the fossil record. We first use genetic parentage information on a large number of wild chimpanzees and mountain gorillas to directly infer their average generation times. We then compare these generation time estimates with those of humans and apply recent estimates of the human mutation rate per generation to derive estimates of split times of great apes and humans that are independent of fossil calibration. We date the human–chimpanzee split to at least 7–8 million years and the population split between Neanderthals and modern humans to 400,000–800,000 y ago. This suggests that molecular divergence dates may not be in conflict with the attribution of 6- to 7-million-y-old fossils to the human lineage and 400,000-y-old fossils to the Neanderthal lineage.



terryt said...

"We date the human–chimpanzee split to at least 7–8 million years"

That will please Maju. I've always been agnostic as to the dating of the split but this date makes sense. Places Sahelanthropus tchadensis right at the base, possibly ancestral to both species. A place where I have long though the fossil belongs:


"Sahelanthropus tchadensis is an extinct hominid species that is dated to about 7 million years ago. Whether it can be regarded as part of the Hominina tree is unclear; there are arguments both supporting and rejecting it. Another complication in its classification is that it is older than the human–chimpanzee divergence (estimated to 6.3 to 5.4 million years ago) seen in genetic data"

Bang goes that last argument. Although the article later says:

"Sahelanthropus may represent a common ancestor of humans and chimpanzees; no consensus has been reached yet by the scientific community".

eurologist said...

A very reasonable estimate - and Maju will be pleased, ;)

sykes.1 said...

So, are we back to the original paleotological bone-based estimate?

Mike Keesey said...

Really great to see an actual independent test of divergence dates. And it fits the fossil record better than the calibrated ones! Extremely cool.

DAE said...

These results are inconsistent with research recently reported on by Science Daily (

"The study, based on the DNA of around 85,000 Icelanders, also calculates the rate of human mutation at high resolution, providing estimates of when human ancestors diverged from nonhuman primates. It is one of two papers published this week by the journal Nature Genetics as well as one published at Nature that shed dramatic new light on human evolution.

"Most mutations come from dad," said David Reich, professor of genetics at Harvard Medical School and a co-leader of the study. In addition to finding 3.3 paternal germline mutations for each maternal mutation, the study also found that the mutation rate in fathers doubles from age 20 to 58 but that there is no association with age in mothers -- a finding that may shed light on conditions, such as autism, that correlate with the father's age.

The study's first author is James Sun, a graduate student in Reich's lab who worked with researchers from deCODE Genetics, a biopharma company based in Reykjavik, Iceland, to analyze about 2,500 short sequences of DNA taken from 85,289 Icelanders in 24,832 father-mother-child trios. The sequences, called microsatellites, vary in the number of times that they repeat, and are known to mutate at a higher rate than average places in the genome.

Reich's team identified 2,058 mutational changes, yielding a rate of mutation that suggests human and chimpanzee ancestral populations diverged between 3.7 million and 6.6 million years ago.

A second team, also based at deCODE Genetics (but not involving HMS researchers), published a paper this week in Nature on a large-scale direct estimate of the rate of single nucleotide substitutions in human genomes (a different type of mutation process), and came to largely consistent findings."