Uncommon Descent Serving The Intelligent Design Community

Another Darwinian “Prediction” Bites the Dust

Share
Facebook
Twitter
LinkedIn
Flipboard
Print
Email

I’m including some quotes from a Science magazine feature that is just out. I just posted the other day about results that conflict with Darwinian expectations/predictions. We’ll just add this to the list. You know, you really have to be a true believer to keep insisting that your idea is correct when almost invariably every true prediction made using your idea turns out to be wrong. I’m beyond even being surprised any more. Here goes:

“It was the most radical of a flurry of recent discoveries of human genes that evolution has strongly favored, a process called positive selection. Four years ago, researchers thought that they would find hundreds of examples in which an advantageous mutation spread rapidly in a particular population. That prediction, based on the first scans of human genome sequence data, did not pan out, and by last year, some researchers were ready to give up.

A growing number of researchers now think it is rare for a particular mutation to spread rapidly to most people within a population, as was the case with the EPAS1 gene.”

The bottom line to all of this is that population genetics is dead. (Have I said that before?!) Notice, though, how the Darwinists talk about it amongst themselves:
‘”In only a handful has there been much progress in identifying the causal mutations and extracting these biological insights about their function,” Sabeti wrote in the 12 February issue of Science (p. 883). Says McVean: “That’s why the whole field—the program of trying to find selective sweeps—kind of ground to a halt.”‘

Never able to admit that they might be wrong, they had to begin all over again in a new and different way.

“Yet McVean and others were convinced that positive selection had shaped much of the genome but lay beneath the radar of methods used to detect it. . . . “It’s very likely that many traits that are different between populations are coded by different alleles; any one may not be so strong,” says population geneticist Pleuni Pennings of Ludwig Maximilian University of Munich in Germany.

Consequently, earlier this year Pritchard and his colleagues proposed an alternative to strong selection on single new mutations. In Current Biology, they argued that selection on more than one gene at once could allow a new trait—such as increased height—to sweep more rapidly through a group.

Detecting such polygenic selection is one of the new frontiers. . . .”

Notice that they are “convinced”. They just know. Isn’t this wonderful?

For those who are interested, we might take a look at whether this makes any sense using simple population genetic metrics. They say polygenic selection works. Let’s see……the mutation rate of the human genome is 1 in 10^8 bp. The human genome is 10^9 bp. So the odds of any single bp changing during an individual human replication is (1 x 10^-9)–the probability of a particular site on the genome changing—times (1 x 10^-8)—the mutational rate = 1 x 10^-17. For the studies they’re conduction, we can assume a population size of 50,000, given the time when this would have occurred and the areas involved. We can assume 8 children born to a couple over a generation time of 20 years; and we can assume that only 60% of the offspring will make it to reproductive age. That means 5 children/2 people over a twenty year period, or 1/8 th of a human per year per member of the population.

So, 50,000 X 1/8 X 10^-8 X 10^-9 = specific bp change/yr = 6.25 X 10-13. For “polygenic” (let’s assume only two—this makes it easy for the Darwinists) “positive selection”, this means the probability of all three sites changing “at once” (6.25 X 10^-9)^2 = approx. 4 X 10^-17 specific bp change/ yr. IOW, for that size population, and this is a very reasonable guess for size, it would take almost twice the life of the universe for them to take place “at once”.

Oh, but, of course, they don’t all have to happen at once. But, wait, they’ve already ruled out “positive selection” as having taken place one gene at a time; so what are we left with except simultaneity? Thus, the invocation of “randomness” in this whole process is pure nonsense. We’re dealing with some kind of programmed response if, in fact, “polygenic selection” is taking place. And, that, of course, means design.

Comments
I was directed to this post today, and was a bit aghast at the numbers I used here. They need to be corrected. For any specific bp, the mutation rate is simply the genomic mutation rate: i.e., 10^-8. We can use a population size of 50,000, that's an OK number. I'm assuming a kind of normalized birth rate=replication rate/year. That's alright. So, the numbers should be roughly: (1x10^8*1/8*5x10^4)^3. The outside exponent is because these 'events' are supposed to happen simultaneously, yet, independently. The probability is, then, (0.625 x 10^-4)^3 = 2.44 x 10^-13 years. Looks like you would need much more time than the age of the universe to arrive at this multiple occurrence. If we assume that one of these three mutations has occurred and become fixed, then we're looking at 3.7 x 10^-9. This is about the entire length of time that life is supposed to have existed here on earth. The only mutation rate that makes sense is for a single mutation to occur: i.e., 625,000 years. None of this looks promising to neo-Darwinism, or drift. It seems the conclusion is sound; but the numbers were way wrong. The corrected numbers, IMO, don't help much as you can see. Please, if the numbers are off, point it out. Thanks.PaV
April 20, 2014
April
04
Apr
20
20
2014
05:33 PM
5
05
33
PM
PDT
As to “standing genetic mutation”, drift accounts for very little mutation over the short history of the human lineage;
Sorry, but you're demonstrating your ignorance again - this doesn't even make sense. The point about standing genetic variation is that it is there, hanging round. Mutations happen, and some drift up to appreciable frequencies (unless selection is strong). This means that there is this variation kicking around: each locus has a small effect, so it is mainly influenced by drift. The expected time to fixation of any allele under drift can be very long, so there are likely to be many drifting genes in any population. You may want to read Mike Lynch's paper that Sal referred to. Although I would suggest you take (and pass) a basic population course first.Heinrich
August 16, 2010
August
08
Aug
16
16
2010
12:19 PM
12
12
19
PM
PDT
Heinrich: I had the stipulation of a "specific" bp, ie. a mutation in a particular spot. If you don't add this factor, which is 1^10-9 (ie, 1 divided by the genome length), then a mutation anywhere is considered acceptable; but, that can't be since the mutation(s) is/are needed at a particular location. As to "standing genetic mutation", drift accounts for very little mutation over the short history of the human lineage; and, besides, the calculation I made assumed tacitly that any mutation at a particular site would be "fixed" immediately. Heinrich, please note that it these kinds of calculations that render Darwinism implausible. I am not opposed to Darwinism because I don't like Englishmen, but because I have never found a satisfactory explanation of how Darwinism (in its neo-Darwinism/modern synthesis) can bring about a change in species---knowing what we know about genomes and genetics. And, if you take the position that somehow the mathematics must be wrong, or that something is missing in the calculations, then you have Behe's book, Edge of Evolution, that demonstrates that, indeed, very little random genomic change ends up finding its way into genes. Lastly, the problem with genomic variation is this: there's TOO MUCH of it! This was the same thing with protein variation when studied in the early 60's, and this led to Kimura's "Neutral Theory".PaV
August 16, 2010
August
08
Aug
16
16
2010
07:38 AM
7
07
38
AM
PDT
I haven't read the article yet, but a couple of comments: 1. McVean's conviction that selection has shaped the genome isn't just faith - there is plentiful evidence at the DNA level for selection having shaped the human genome. 2. The human genomics crowd have been finding that for most traits they can find genetic variation, but finding the genes has been more difficult. There are some obvious statistical reasons for this. the point is that it looks like genetic variation is often due to several genes, and a lot of it could be the effects of rare alleles (either at the same or different loci). 3. PaV's calculation based on mutation rates ignores the possibility that there is standing genetic variation that selection can act on. It is well established theoretically that standing genetic variation will be present, both due to drift and mutation. BTW, the odds of any one bp changing is 1 in 10^8, not 1 in 10^17, so the calculation is out by a factor of 10^16. Or would be if 6.25 X 10^-13 hadn't somehow mutated into 6.25 X 10^-9.Heinrich
August 14, 2010
August
08
Aug
14
14
2010
12:09 PM
12
12
09
PM
PDT
Thanks to those who provided the link. PaVPaV
August 14, 2010
August
08
Aug
14
14
2010
02:54 AM
2
02
54
AM
PDT
Tracing Evolution's Recent Fingerprints Ann Gibbons Science 13 August 2010: Vol. 329. no. 5993, pp. 740 - 742 DOI: 10.1126/science.329.5993.740 http://www.sciencemag.org/cgi/content/summary/329/5993/740Enezio E. De Almeida Filho
August 13, 2010
August
08
Aug
13
13
2010
05:37 PM
5
05
37
PM
PDT
The discovery of a gene that helps Tibetans thrive at high altitudes is the most radical of a flurry of recent discoveries of human genes that evolution has strongly favored, a process called positive selection. Four years ago, researchers thought that they would find hundreds of examples in which an advantageous mutation spread rapidly in a particular population. That prediction, based on the first scans of human genome sequence data, did not pan out, and by last year, some researchers were ready to give up. [b]They now realize, however, that plenty of positive selection exists, but it is subtler and harder to trace than originally anticipated.[/b] Using new statistical methods, they have found many less dramatic mutations that, for example, also help highland Tibetans survive at high altitude. Others allow Yupik Eskimos to stay warm efficiently, Europeans to thrive on cereal grains, and, perhaps, East Asians avoid alcoholism.
(some omitted text restored) Link: http://www.sciencemag.org/cgi/content/summary/sci;329/5993/740?maxtoshow=&hits=10&RESULTFORMAT=&fulltext=polygenic+selection&searchid=1&FIRSTINDEX=0&sortspec=date&resourcetype=HWCITPetrushka
August 13, 2010
August
08
Aug
13
13
2010
03:59 PM
3
03
59
PM
PDT
PaV, Would you kindly provide a link to the Science article?Adel DiBagno
August 13, 2010
August
08
Aug
13
13
2010
01:47 PM
1
01
47
PM
PDT
avocationist
Does anyone know if someone has compiled a list, an article or a book that deals with the subject of failed Darwinian predictions? This would be a great project and one I would love to read.
Don't know about a book but Cornelius Hunter has a site on that: http://www.darwinspredictions.com/Borne
August 13, 2010
August
08
Aug
13
13
2010
01:25 PM
1
01
25
PM
PDT
Does anyone know if someone has compiled a list, an article or a book that deals with the subject of failed Darwinian predictions? This would be a great project and one I would love to read.avocationist
August 13, 2010
August
08
Aug
13
13
2010
12:44 PM
12
12
44
PM
PDT

Leave a Reply