Comments on "Heliocentric Certainty Against a Bottleneck of Two"?

(S. Joshua Swamidass) #1

Continuing the discussion from Heliocentric Certainty Against a Bottleneck of Two?:

Please include comments and questions to the original post here.

Heliocentric Certainty Against a Bottleneck of Two?
Heliocentric Certainty Against a Bottleneck of Two?
(S. Joshua Swamidass) #2

21 posts were split to a new topic: Burke’s Objections to Heliocentric Certainty Thread

(Sy Garte) #4


First I need to apologize for being slow in answering, and also for not being able at the moment to take the time needed to give a careful and well thought out review of your work. We are all busy of course, and my hope that retirement would make me a man of leisure has long been dashed, but this particular period has been (and will continue to be) pretty intense. More about that later at some other time.

Given my time constraints, and the rustiness of my pop gen skills (its been a while since I’ve used them) I have not checked your work in detail, but it seems to me to be generally accurate, or at least to make sense. As you indicated somewhere, (I cant find it at the moment) the population size of a bottleneck is inversely proportional to the length of time between the bottleneck and the present, given a particular degree of genetic diversity in the present. Thus, your result that a possible bottleneck of 2 about 500Kya seems plausible. And of course, we know that despite Dennis’ comment about modern human genetic diversity, humans are far less diverse than other, older primates, based on Wright’s Fst statistic, which is something like 2, as compared to about 10 for gorillas, if I recall correctly. So, based on this alternative approach, it seems that your result is certainly feasible.

Of course, what that explains is the potential for a bottleneck of 2 to allow for modern genetic diversity, which we all agree could not be possible if the bottleneck happened after 100Kya. The figure of 10,000 (I think 2000 to 10,000 is still used by some) cited by Dennis is likely the result of a bottleneck believed to have occurred as a result of the general die off following the Toba explosion around 70Kya that produced a global cooling and drying trend. But this says nothing about the possibility of earlier bottlenecks. So if the quest for a bottleneck of 2 is the goal, I believe you have found it.

The real problem for me, though is the theological implications. At 500Kya, no couple could have been anything like Adam and Eve, of course. They would not have been H. Sapiens, its likely they didn’t speak, probably couldn’t have worshipped God, their immediate descendants certainly could not have tilled the land or built cities, etc. The extent of their humanness, their possession of anything resembling Imago Dei, must be considered very limited.

And the issue also remains that theologically we cannot speak of bottlenecks so much as special creation. That is still possible with a putative common and restricted ancestry of 2 individuals, but quite frankly I don’t see the point. Even if your calculations are correct (and I believe they are), what you have shown is that all current human beings are descended from one and only one particular couple that lived before modern H Sapiens were a thing. And even if that couple were the result of a special creation, their immediate descendants still had to mate with the surrounding population. And then allow a further 200-300K years to pass for the evolution of H Sapiens to occur.

I find the genealogical argument that we are all descended from many couples who lived 6 to 10,000 ya, one of whom were Adam and Eve, people either created by God or endowed with spiritual gifts, to be far more compelling. I realize that some will reject the many couple idea, but they cannot use a literal reading of scripture to do so. The single creation event is interpretation, and as you know, I reject it.

So in summary, while I believe that your attempt to find a single universal MRCA of 2 people has probably succeeded, I find less value in that though, than I do in your earlier discovery of the genealogical common ancestry of all people in an historical Adam and Eve within a Biblical time frame.

(S. Joshua Swamidass) #7

Thanks for your comments @sygarte.

It is not so much my “quest” or “goal.” I’m just trying to understand what the evidence does and does not tell us. That is it.

How do you know this? As I understand it, this is a matter of great debate among anthropologists. In the last couple decades, as I understand it, the pendulum has swung away from this view. I think the fact of the matter is that we do not really know how “human” Neanderthals and other hominids really were. The only agreed upon distinction appears to be that they failed to survive to the point that Homo sapiens civilization arises. That does not tell us much about their intrinsic abilities.

I do affirm that some scientists hold your view. Others do not. I think we just do not really know.

I agree. A Genealogical Adam (recent) seems to be much stronger from a theological point of view. Stil there is value scientifically in understanding the evidence.

It all depends what we mean by “special creation”. I won’t get into the miraculous origins models in detail now, because distracts from the scientific piece. However, but take the Biola belief statement as an example.

God specially created Adam and Eve (Adam’s body from non-living material, and his spiritual nature immediately from God).

If God created Adam’s body by reanimating a dead hominid (non-living material), that would create genetic continuity between Adam and prior beings, and leave the impression of common descent. I’m not at all advocating this position, but I am offering it as a view that some may feel compelled to take. Is this special creation or common descent? Depending on how one looks at it, we could see it as either. Evidence, however, does not seem to adjudicate one way or another.

Incidentally @sygarte, will you be at the 2018 ASA conference? I’m doing a workshop on the science of Adam. Would you be interested in participating?

(Sy Garte) #13

Yes, Im pretty sure I will be there. I might be interested in participating in the workshop, but not sure what I could contribute. Let’s talk in about a month (some things will be clearer then).

(Mark M Moore) #25

That is a lot of very impressive material to digest. As you may remember, I don’t think there was ever a bottleneck of two, but I ask what is next out of curiosity. Many if not most scientists in what I guess I should call genetic anthropology believe that there was limited but detectable admixture with non-humans around 55,000 years ago. For example my genetic testing identified 1.5% of my genome as “Neanderthal”. East Asians tend to have marginally higher amounts. Sub-Saharan Africans much lower. And a few isolated groups have even higher contributions from another hominid. In each case, apparently the percentages were higher in the past, and mostly somewhat deleterious as evolution is weeding out most of the non-human “contribution” to Homo Sapiens over time.

My question is this, wouldn’t this set of genes, which are different enough from the human versions to show a result of much greater than 495K years for a “common ancestor” gene, skew the results of your calculations older? I mean if it measured a group of 50 genes and 48 of them were calculated to have a MRCA of 150K ago but the other two - which were introgressed from non-humans- had a MRCA of 1.5 Million years ago then wouldn’t the average be driven upward in a misleading fashion by the limited contribution of the archaic material?

(S. Joshua Swamidass) #26

In general, yes. That is why we used a median estimate of TMR4A instead of (for example) the max or mean. The median estimate is only going to minimally affected by this. If 10% of the genome is subject to interbreeding artifacts, for example, we’d only expect that would increase the estimate by about 20 kya. So no this ends up not being a large source of error.

(S. Joshua Swamidass) #27

At the moment, there are edits being privately proposed to this statement.

How do you think it should be written?

(Mark M Moore) #28

I’d write it mostly the way you wrote it since that appears to be what the numbers say. I would change this sentence

Is the assumption “that humans share common ancestry with apes,” even strictly necessary to determine your numbers since you are comparing diversity in humans with the rate of mutation in humans? Even if it was a part of the calculations of some of the research which you drew data from, I see it as superfluous and a potential distraction from your actual findings. I don’t see where it is essential that your team assume that in order to get the numbers that you did.

Regarding the phrase “God has not intervened with physical miracles” I think I would be way more specific. Something like “We assume God has not intervened in a super-natural way so as to give a false appearance of age to the genetic data.” I think your results are going to be dismissed as based on invalid assumptions by those who believe in miracles unless you phrase it in such a way that you are only assuming He won’t do any miracles which are outside His character.

The only thing else I might suggest, and maybe this is the next study and not this one, is that you address the idea that a flood wiped out all Y-chromosomes except for one sometime long after Adam. If your MRCA for just Y-chromosomes also shows a date in the distant past, it might be worthwhile to point that out. There is not just the question of the nature of Adam to consider, but the flood and Noah as well.

(Vincent Torley) #29

Hi Joshua,

By and large, I’m pretty happy with the statement. Regarding the controverted sentence, “We have assumed that humans share common ancestry with apes, and that God has not intervened with physical miracles,” I’d write: “We have assumed that humans share common ancestry with apes, and we assume by default that God has not supernaturally altered our DNA since the time when the line leading to humans diverged from the line leading to the apes that are most closely related to us.” How does that sound?

(Mark M Moore) #30

@swamidass this comment that I am replying to made on Feb. 11th, and your response below it.

(S. Joshua Swamidass) #31

I suppose this is my hypothesis, and we can test it for sure soon.

Remember, I’ve been arguing all along that:

Though, at this time, I did find one study.

(George) #32

ORIGINAL VERSION by @swamidass:
“We have assumed that humans share common ancestry with apes, and that God has not…
… intervened with physical miracles,”

gbrooks9 VERSION:
““We have assumed that humans share common ancestry with apes, and we assume by default that God
[does] not …
… use miracles to override natural processes for creating needed mutations only when nature alone will be tardy in producing the necessary result, whether tardy by a minute, a year, or not ever."

This makes for quite the “spread” in meanings!

(Mark M Moore) #33

I feel pretty confident that it is not much more than 10% unless you are measuring Paupuans or something with lots of Denisovan.

Here are some factoids I remember which might apply to this problem and put bounds on it. 1) The total percentage of a Neanderthal Genome which can be found in all human genomes is 20%. Many of those though, may be the ancestral gene common to human populations lost in the humans that survived so you decide where those go on the ledger. 2) East Asians tend to have a number of DIFFERENT gene points than the rest of Eurasians, so that if Europeans like me have 1.5% and a man from China has 2% Neanderthal genes the real number to use would not be the average of those two numbers but something above that closer to but short of the total of those two numbers. 3) The neanderthal genes are not random, some regions are very depleted especially on the Y-Chromosome. 4) The total number of breeding events between humans and Neanderthals is estimated to have been very low. I saw one estimate that said it was dozens of times, not hundreds or thousands! For even 1.5% of those genes to be retained the initial OOA population must have been quite low.

(S. Joshua Swamidass) #34

Well, we will find out soon enough.

(Joseph Akins) #35

Thank you for this discussion and the related threads; very helpful. This may have already been covered and I missed it, but allow me to play “skeptic’s advocate” on a different aspect of the general heliocentricity comparison:

The notion that we have the same certainty of past population structures of Homo Sapiens Sapiens (HSS) as that of heliocenticity is at once an homage to Icarus and an insult to Galileo. Where are the equivalents to moons orbiting Jupiter and phases of Venus that justify equivalent certainty in inferring historical breeding population sizes?

What the scientifically literate person on the street knows with equal certainty to heliocentrism regarding evolution is survival of the fittest. The faster lion can catch more prey at once increasing its own reproductive success and culling weaker genes from the population of zebras. However, such objectively reproducible observations that we can experience on a daily basis (on TV, or in person) are irrelevant to the discussion of inferring the probability that all existing humans descend from a genetically isolated divinely created couple ~200ya?

Ok, what objectively verifiable observations support the notion that the population of HSS has never dropped below 1,000? Although coalescent theory is designed to infer past gene genealogies mathematically, where is the objective empirical evidence that we can trust population models to reflect historical genetic variation? Where has the WF, or any other population growth, model been used to predict genetic variation in a population before such variation has been measured? Why couldn’t it be that as HSS migrated out of NE Africa they encountered new environmental stresses that led to accelerated rates of evolution? Assuming the mutations are neutral, especially in genes related to immune response, needs to be justified.

Having done this, it would be very useful to not only demonstrate the success of Neutral Theory to constrain past population structures independent of mutations for certain loci (which you have done), but to show plots of the needed 1st and 2nd derivatives of the mutation rate for adaptively driven loci. I’m not sure if the previous sentence makes sense, but basically I don’t think most people, Creationist or not, get your “drift”.

(S. Joshua Swamidass) #36

This is not about natural selection. Rather we are talking about population genetics. None of this depends on common descent, but on very widely agreed upon (even among creationists!) knowledge about how DNA changes over time.

Did you read the main thread yet? (Heliocentric Certainty Against a Bottleneck of Two?) You’d see that we all agree that there are no HSS around at 300,000 years ago, so the population certainly dropped down below 1,000. HSS are not our only ancestors, and they had non HSS ancestors too. Right?

This however is a very weak and essentially fallacious argument. Most of the genome is neutral. Immune related genes? We are not relying them for the inference. The median TMR4A avoids the outliers. So there is very good reason that our ancestors do not drop to a single couple before 500 kya, and probably not before 700 kya, even though there are precisely zero HSS at this time.

It would be better to just ignore the adaptive loci. In the vast majority of cases, adaptive loci look younger than the rest of the genome. Immune sequences look older but are more rare. Focusing in on the neutral areas would most likely increase the TMR4A, but probably not by much.

The main thread is highly technical and meant to demonstrate the case to other scientists, not for the public so much. For the general public, point them here:

(S. Joshua Swamidass) #37

I agree. That very unhelpful rhetoric.

(Joseph Akins) #38

I was presenting an amalgamation of protests to population genetics inferences regarding Adam and Eve fishing for some clear cut rebuttals that I’m not aware of. With the lack of exposure to neutral theory and population genetics among scientists, nevermind “the man on the street”, it seems you are simply appealing to voodoo math. I understand that you are not, but again a significant portion of thinking about evolution is still preoccupied with panadaptationism. The paradigm shift to Neutral Theory is only in its infancy culturally, so some sensitivity to guiding the shift from the Ivory Tower would be useful though not necessarily required.