Question: Why is this an accurate simulation for haplotype distributions? A Gaussian distribution is essentially pure noise except for the mean and the variance. @DennisVenema is claiming that the existence of TMRCA values > mya is a legitimate signal of ancestry, not just noise. If the observed PDF across the genome is the result of a stochastic process that makes some regions seem younger than the ground truth (as measured by TMRCA), then the noise should not be interpreted as nullifying the signal from regions at extrema that point to a very ancient TMRCA.
I donāt know what you mean by correcting for interbreeding here. If many of us are descendants of ancient interbreeding between lineages that diverged at least half a million years ago (which we are), then we are not a product of a bottleneck of size two within the last half million years. To take seriously a recent tight bottleneck, you donāt have to correct for introgression ā you have to ignore it.
Iāll post more later. This is just regarding @DennisVenemaās claim re: Homo sapiens. If we care about measuring Homo sapien population size, we have to correct for interbreeding.
The good news is that I found exactly the data we need to do this study. It includes a scan of phylogenies constructed across the entire autosomal genome. Each phylogeny is computed for non-recombining blocks, which corrects for recombination introduced artifacts. From this we can compute TMRCA and T4MRCA across the whole genome, by looking at the times of the first and third coalescent nodes. Once we rescale to years with the mutation rates, we will have the distribution across the whole genome.
We can test to see how well TMRCA / 4 = TMR4A.
We can see the distribution of TMR4A across the genome.
We can identify the outlier areas, or subset on any region of the genome.
The bad news is that the source data is 424 GB of compressed data. I got a plan to handle it though. If that is a good analysis to do?
I have to admit that I havenāt looked at the precise wording of Dennisās statements lately, but my impression was that his claims were not about āorganisms arbitrarily labelled as Homo sapiensā but about āusā, which would correspond to contemporary Homo sapiens. Contemporary Homo sapiens did not go through a bottleneck of size two 200,000 years ago. My ancestors did not go through a tight bottleneck 200,000 years ago. At that time, some of them were living in Africa and some of them were living in Europe. I still donāt see why the demographic history of one branch of the structured population I descend from has anything other than purely academic interest to anyone. The fact that many (but not all) biologists have slapped different names on the different branches doesnāt strike me as relevant at all.
I once pointed out in another group that if we go back far enough, both sides of all human lineage is African!
I thought that was a pretty clever turn of the phrase. But wouldnāt you know I was soon confronted by a Texan, amateur anthropologist and white supremacist, who insisted that some human races have no ancestors from Africa.
Yeah, that was the quotation I was basing my response on; I exploited your effort without citing it (or even liking it ā bad glipsnort). What I donāt know is whether Dennis wrote other things that phrased the claim in different ways, and to which this discussion would actually be relevant.
I went through the first 290 posts looking for any relevant statements by @DennisVenema. I donāt see any that can be construed as saying that a bottleneck to two in Homo Sapiens (as opposed to āour lineageā including interbreeding) is ruled out with certainty comparable to heliocentrism. This is getting pretty nitpicky, but if it aids the discussion then good.
I fully agree with you and Iāve raised similar points about peopleās tone in other threads, only to be told otherwise by moderators. Obviously, itās nearly impossible to police for tone leaving only overtly hostile or crude comments as objectively inappropriate. However, my guess is that moderators who frequently see and censor that sort of dialogue end up with overly low expectations of what civil dialogue looks like. Unfortunately, to non-experts reading these sorts of exchanges, tone plays a large role because itās one of the few elements that they can actually assess. So while itās not really realistic to moderate, I think that we could at least admit that it could often be better and that as Christian brothers and sisters, we should strive for the most gracious dialogue possible. I think that Dennis (and all Biologos writers, really), are completely above board on this and I honestly think that that is one of the most persuasive aspects of their position.
While I have trouble what post of tallen_1 you are referring to, agree with your comments on tone. As a suggestion, if you comment on a post that is way back in the thread, highlight and quote it to allow others to easily see what you are referring to.
Part of the problem is that some seem to see these discussions as competitions to be won by the knights carrying their colors, when they are more an effort to reach understanding of the relative positions.
Itās a bit odd to see this dug back up, but my concern now is simply one of basic fairness rather than tone. By now weāre all accustomed to Richardās inquisitorial conversation style. It is what it is. What I do want to see happen, and what I think Richard owes his conversational partners on this thread, is to abide by his same standards of accountability in openly and honestly answering questions posed to one about what the evidence does or does not support. As long as he plays by the same rules as he holds others to, then I think weāre good. But if he thinks the only important questions to answer are the ones he pushes, then I donāt think we are. He has some unanswered pertinent questions put to him by Dennis and others. Letās see what he does with them.
I have not read every single post, but I have read with interest the various expert opinions, and for what it is worth, I give my impression.
The certainty that is displayed by one side seems to be the absence of modelling that would support a ābottleneckā of two (I find it hard to believe two creatures are a bottleneck? but this is the terminology).
The questions that are asked however point to a ālack of similar certaintyā regarding what I think are important parameters in the modelling.
I (as a non-participant) am at a loss as to the physical evidence that is supposed regarding this bottleneck of a few hundred or whatever number is discussed. Am I missing something? Has someone discovered remains at a particular location to show such a group existed at some point in time? If not, is this not an inference derived from the modelling. If there is direct physical evidence, why is this not discussed at length, as it would be relevant to the certainty/uncertainty aspect of the discussion?
Yes - one can use considerable expertise and ingenuity to argue whether the population genetics model supports a particular contention or not. Thatās the nature of this thread.
But the question of whether the model itself is adequately valid over such timescales, given its known limitations and the state of flux of theories of large-scale evolution, is a significant one.
āAll models are wrong - some are usefulā. But their utility is only measurable by the ability to validate them by independent observation under the situation for which they are being used - in this case the origin of humanity defined, at least, as our species or even across hominin species by some protagonists. Thatās very different from studying the evolution of Y-chromosomes in the living population.
In this case, validation would seem to require counting fossils that are as rare as hensā teeth - in the absence of physical evidence, the population genetics model seems to validate itself in a circular manner.
Good question, George. Assuming I have understood this thread, the data being predicted/compared with current state and the data used in rate calculations are not the same. The predictions/observations of current state focus on the distribution of genomic features, whereas the assumptions of rates of change are based on known history.
Validation of models, as I have practiced (and is commonly understood) requires a result from the model to be similar to an observation/measurement of a system independent of the model data base. Within this I have a difficult time noting such validation of the models discussed here. Since a major point is the size and time of a bottleneck, for example, validation would be considered by using physical data of a real bottleneck. There may be other ways, and if you can identify them I would be interested to know.
This is not a question on the technicalities of the modelling proceure.
