Comparisons of Neandertal mtDNA with modern humans and modern chimpanzees

Several threads have been discussing research from the late nineties, which involved the comparison of mitochondrial DNA extracted from Neanderthal remains with mtDNA in humans, and in chimpanzees, as a side issue from the main thread topic. This thread is intended to allow focussed discussion of the mtDNA comparisons.Two papers have been cited in this regard, and both are on-line.

1. Neandertal DNA Sequences and the Origin of Modern Humans
   by M. Krings, A Stone, R. W. Schmitz, H. Krainitzki, M. Stoneking, S. Pbo
   in Cell, Vol. 90, pp 19—30, July 11, 1997.
2. DNA sequence of the mitochondrial hypervariable region II from the Neandertal type specimen
   by M. Krings, H. Geisert, R. W. Schmitz, H. Krainitzki, S. Pbo
   in PNAS Vol. 96, Issue 10, pp 5581-5585, May 11, 1999

Some commentary on the first paper in particular has erroneously described the results as saying that Neandertal mtDNA is about halfway between ours and that of a chimpanzee. However, none of the scientists involved in the research say anything of the kind, and the empirical data flatly contradicts that assertion.It is, on the other hand, a natural error to make by a novice reading the first paper[1], since the raw data was not shown, and the following diagram does seem to suggest some kind of halfway aspect to the data.
(Figure 6 from Krings et. al. (1997); click for full size)To better show what is going on, I have produced another parallel diagram for comparison, based on the second paper. The diagram above omits the Neandertal-chimp differences, and the paper does not report those comparisons, although I understand that they were done. The second paper refers to a second independent experiment, involving a slightly longer sequence of mtDNA and more details on the comparison data. In this case the data set involved 663 human mtDNA sequences, 1 Neandertal sequence, 7 common chimpanzee sequences, and 2 pygmy chimpanzee sequences (bonobos), and every sequences was compared with every other sequence.The pairwise differences can be plotted as follows (click on image for full size in a new window):
This is not a direct plot of the data set, since I do not have that available. Instead, I plotted normal curves based on the quoted mean and standard deviation for each class of comparison. Because there are two species of chimpanzee, the Chimp-Chimp comparison has a wide spread of values. Because there is only one Neandertal sequence, there is no Neandertal-Neandertal comparison.The critical point to note here is that Neandertals and Humans are almost exactly the same distance from Chimpanzees. This is exactly what we should expect from an evolutionary perspective, since biologically speaking, Neandertals and humans are both equally evolved from chimpanzees. This result is a contrast to the common misunderstanding of evolution as a kind of ladder of progress, with some species being less evolved than others, or with one existing species between "between" another two existing species.Cheers -- Sylas(Added in edit. The plot has been altered since posting the article. If you don't see a curve for the differences between common chimpanzee and bonobos, then refresh your browser. Images reduced, linked. Caution... some links open in your current window.)[This message has been edited AdminSylas, 05-03-2004]

That is true, on both counts. Note that the method I used to produce the plot was to enter mean and standard deviations quoted from Table 1 of Krings et. Al. (1999) into Excel, and then produce normal distributions to let them be more easily visualised. The diagram cautions that it is not a direct plot, but a crude representation of the mean and spread of pairwise differences in the given class.That paper does also give some information which allows us to put the Human-Neandertal spread of differences into perspective, by comparison with the subspecies of common chimpanzee.From Neandertal DNA Sequences and the Origin of Modern Humans
by M. Krings, A Stone, R. W. Schmitz, H. Krainitzki, M. Stoneking, S. Pbo
in Cell, Vol. 90, pp 19—30, July 11, 1997. We can illustrate this paragraph by the same method as I used for the previous plot. I enter the mean and standard deviations quoted into excel, and then produce a plot of normal distributions with those parameters. This gives a crude visual representation of the relative differences. (Click on the image to see it full size in a new window.)
There is no plot of the diversity within humans, which is unfortunate. I expect it would be less than the diversity within any chimpanzee subspecies. We can't compare directly with the numbers for the HVRII.There is no indication of the difference between bonobos and any subspecies of common chimpanzee, but this would certainly be greater than the other differences plotted.Unfortunately, I don't have the raw data. I would love to get a complete list of pairwise differences, or the raw sequence data. I think I will be able to do so, but if anyone can beat me to it and make the data available easily, I'll be grateful. Sequence data is available, but it is a lot of work to track down the various queries and put it all together.The data suggests that Neandertals ought to be regarded as a subspecies, not a completely distinct species. The degree of difference is sufficient to make them a clearly identifiable distinct group, in contrast to any existing human races, where differences between races are much closer in magnitude to individual differences within races. The conclusion that Neandertals and modern humans had a long and distinct history with little if any gene flow between the populations is strongly supported. The conclusion that they were completely distinct species is weak. They would certainly have been interfertile with each other, but the extent of actual cross breeding was, as far as we can tell, minimal and had little if any contribution to the present human gene pool.The sequence data entered into Genbank by Krings et al (accession number AF011222) identifies the source organism by a subspecies designation: Homo sapiens neandertalensis. Some authorities prefer a species designation Homo neandertalensis. For extinct groupings, this distinction takes on a rather subjective character.Cheers -- Sylas(Edited to reduce image size)[This message has been edited Sylas, 05-03-2004]

No; that was the HVRII. There was no difference given for the HVRI, which was the subject of my second plot. We could add the 75.7 difference to the first plot, which was based on HVRII; but I omitted it first time round as I was working from data in table 1. I will add it in for comparison to the first plot, by modifying the deep linked diagram in Message 1. I'll indicate when it is complete by a comment in the post; after which readers may need to refresh the browser to get the new image. I'm collecting a set of references for this. Can you give a cite where I could track down more on these, please? Does mtDNA diversity give any different result to nuclear DNA diversity?One reference I have is A view of Neandertal genetic diversity by Krings et. al. in Nature Genetics (brief communications) pp 145-146, vol 26, Oct 2000, but it lumps chimpanzees as one grouping, with consequent high diversity. Yes, I should not have said "certainly". However, there is no actual evidential basis for inferring lack of interfertility, and the close similarity on morphological and genetic grounds makes it very implausible. I'm not saying that there was any great amount of gene flow; available evidence suggests this was minimal to non-existent. But that is a different matter to lack of interfertility.Cheers -- Sylas

Of course it isn't critical. Your case is rather different. That is one of the things I have been trying to tell you, and why it is so weird that you can't acknowledge the error in the description you have quoted from the Indian newspaper.Rather than fix up your presentation to match the actual data, you have continued to insist on the reporter's erroneous description, and even insisted that this phrasing is repeated by the scientists involved. That's false as well, as we saw with the Google searches in the other thread.Why is it so hard for you to fix such straightforward errors? The first phrase is true, and the second is false. The findings do not give any indication that interbreeding was precluded. They only indicate that there was no detectable genetic contribution made. These are different claims.The paper indicates that the difference between Humans and Neandertals is less than the difference between Western and Eastern varieties of the common chimpanzees... and they are completely interfertile, and yet also with distinct gene pools because interbreeding does not usually occur. There is a hilarious irony here in you speaking of ignoring deductions, given that you've had YEARS to fix the halfway error.But in any case, the above paragraph is very silly. The reason we are not descended from Neandertals is because we already know that Homo sapiens lived at the same time as the Neandertals.The question is not really ancestry it is whether or not there was interbreeding. The data had the potential to show Neandertal genetic contributions to the current human gene pool. Even if this had been the case, we already know that humans Homo sapiens have a fairly continuous lineage distinct from the Neandertals. What was at issue was the multi-regional hypothesis, and the question of whether, when Homo sapiens moved into those regions where the Neandertals were living, there was much gene flow between the populations.The data suggests that there was not. All the various human "races" are pretty much equally distant from Neandertals. If there was interbreeding, we should expect to find people in Europe to be a bit closer to the Neandertals. But they aren't. The genetic distance indicates a common ancestor around several hundred thousand years ago, whereas we lived side by side with Neandertals about 40,000 years ago.If Neandertals remains were all dated to several hundred thousand years ago, there would be no particular problem with ancestry. Ancestry is emphatically not ruled out just on difference. You need to consider timing.Rest ignored; it's just based on the error that excessive difference alone rules out ancestry. That's incorrect.Cheers -- Sylas[This message has been edited Sylas, 05-02-2004]

More to the point, we do have the remains of such hominids; many of them. Redwolf even mentions some, but his discussion is flawed by various other errors. I'm sure we'll discuss this some more as the thread continues. Neadertal remains have been found more a bit more widely than this. As well as finds in Europe, there are finds in Syria, Israel, Uzbekistan, Morocco and Iraq. Some of these fossils are described at Homo neanderthalensis, at Stephen Heslip's pages at Michigan State University, for a course on Hominid Fossils.Cheers -- Sylas

Not that I know of. The find at Jebel Irhoud, which is the basis for my listing Morroco in the previous post, would be Africa. It is just across from Gibraltar, but I think the identification with Neandertal may be contentious.The Neandertal range is given by Francis Steen as follows:
{Image is a link to source; uses current browser window}

The source for this map is the same that redwolf used in Message 9 for skull pictures. Seems to be comprehensive and high quality, with copious references to diverse views within the scientific community. Thanks for the reference.Cheers -- Sylas Edit change. The map was previously taken from Neanderthal Sites (by D.S. McDonald). It is now taken from a different source, used by redwolf, and is properly linked.[This message has been edited Sylas, 05-03-2004]

I read it better than you, apparently! The very next paragraph after the bit redwolf gives in red reads as follows (my emphasis in bold). Did you read that bit?Yes, I read your link, and the other ones you gave also. You can get all kinds of diverse opinions on this. I have no problem if it turns out that Shreeve disagrees with me, and indeed no problem if there is new evidence that leads me to change my mind. Interbreeding, and interfertility, are still open questions, and the data continues to be collected and examined and tested in various new ways. Shreeve is explicit about this in the end of the article you have quoted: Here are a couple more points you should note.

• Shreeve is a well regarded science writer, but not actually a scientist. He has written some interesting books on Neanderthals, which have been widely praised; but they are not definitive works by any means. They review work by others, and present his own speculations on the matter.
  ​
• Shreeve's own speculations on the lack of interbreeding are that the two species may well have been interfertile, but that interbreeding did not occur because of different mate recognition signals. The argument for this is entirely circumstantial, but worthy of serious consideration. Shreeve himself recognizes this as speculation.
  ​
• The bit you placed in red is not Shreeve's own view, but the way Shreeve describes the views of unnamed German scientists. Shreeve gives no citation and does not name the scientists, but most likely he is referring to Krings et al, whom we have discussed here at some length. The clue is that in the previous page of your link, Shreeve refers to DNA extracted from an arm bone; this is just what Krings et al did. In this case, Shreeve is wrong. The bit that he gives in red does not actually correspond to the views expressed by Krings et al. If you disagree, give a quote from the primary sources. The links are available in Message 1. But don't bother. The truth is that the scientists involved limited their conclusions to the notion that there was no genetic contribution to the gene pool; not that the differences were too great for breeding to be possible. In fact, they gave the comparison with chimpanzee subspecies, suggestive that they were interfertile, even though little to no breeding took place.
  ​
• In your second link, Shreeve does not make a definitive case, but a speculation, by his own description. Quoting from your second link... (my emphasis):
  Although it is merely a speculation, the idea fits some of the facts and solves some of the problems. Certainly the Neanderthals' ancestors were geographically cut off from other populations enough to allow some new mate-recognition system to emerge. During glacial periods, contact through Asia was blocked by the polar glaciers and vast uninhabitable tundra. Mountain glaciers between the Black and Caspian Seas all but completed a barrier to the south. "The Neanderthals are a textbook case for how to get a separate species," archeologist John Shea told me. "Isolate them for 100,000 years, then melt the glaciers and let 'em loose."

  ​

  If mate recognition lay behind a species-level difference between Neanderthals and moderns, the Levantine paradox can finally be put to rest. Their cohabitation with moderns no longer needs explanation. Neanderthals and moderns managed to coexist through long millennia, doing the same humanlike things but without interbreeding, simply because the issue never really came up.
  That is, Shreeve disagrees with your extreme position. He considers that the reason for lack of interbreeding was not lack of fertility, but lack of desire.
  ​
  In fact, your second link is a rather beautiful and thought provoking speculation. I enjoyed it, and it is intriguing to wonder what human history would have been like if our Neandertal cousins had survived.

In the meantime, my extract quoted above remains unambiguously true; it is a straight description of the actual data which we need to consider. The actual empirical data indicates that the genetic difference between Neandertals and Humans was less than the difference between interfertile subspecies of the common chimpanzee, and much less than the difference between common chimpanzee and bonobos, which are also interfertile.Indeed Shreeve mentions this himself, in your first linked article!Shreeve's speculation (not analysis) is an attempt to reconcile the evidence I have cited on comparisons with chimpanzee diversity (which indicated that Neadertals probably were interfertile with Homo sapiens) and the genetic evidence of Krings et al (which suggests that Neadertals did not contribute to the modern gene pool). And even then, Shreeve acknowledges that there is room for future evidence to contribute further understanding. I agree with him on all these three points; and I think his proposed solution, of lack of mate recognition, is very plausible.I think you need to read that article again, and more carefully.Cheers -- Sylas PS. You would be better to remove "?term=" from your link to the first Shreeve article. I have removed it from the URL in the quoted extract from your post. Link construction code here fails to parse the "=" as part of the link. You can also work around the problem by making the links explicitly, rather than leaving a bare URL in your post.

Sheesh -- get your priorities right. You should always have a Buffy episode on in another window. I have another question, which you or others may be able to help me with. In previous posts, I have assumed that the chimpanzee subspecies are interfertile. However, I can't confirm this; and on checking I could be wrong.It seems that the close similarity indicates to most researchers that hybrids are extremely likely, and if I recall correctly, conservation groups advocate that chimpanzees in captivity should be maintained in their distinct subspecies in order to prevent cross breeding and loss of diversity.Is anyone aware of whether or not there are any confirmed instances of fertile hybrid offspring from any of the chimpanzee subspecies? Are there any populations of distinct subspecies which are in contact with each other, so that interactions across subspecies lines would be possible?Cheers -- Sylas

(Added in edit... fixed the source to be "Sylas") This is a very good point, and your post gives some well known cautionary tales.There has been an attempt to test for some of the problems you mention. For example Extract from the paper: This paper, by the way, is a comparison of Neandertals with contemporaneous modern humans (Cro Magnon). The conclusions are summarized in figure 2 of the paper. The following image is a link to the source, which is given in a new window. The associated description is quoted from the paper. "Average genetic distance between ancient and modern samples (2,566 sequences of modern Europeans; y axis), as a function of the samples' age (x axis, in thousands of years). Vertical lines represent two standard deviations above and below the mean. Squares, anatomically modern humans. Diamonds, Neandertals. The Paglicci samples typed in this study are indicated by open squares. The point at 0 years indicates the average pairwise difference between present-day samples."
The diagram shows a sudden discontinuity for Neandertals, and a consistency for anatomically modern humans which indicates that the methods for handling ancient DNA are probably reliable.Cheers -- SylasThis message has been edited by Sylas, 05-04-2004 09:53 PM