Clearly @kleinman believes that 90m/100m=90m, and that we are the ones bad at math.
1 Like
This is a very basic concept. Has @kleinman acknowledged it yet?
2 Likes
Nope. Acknowledging isnât really his thing.
1 Like
I provided the probabilities you asked for and showed how you were wrong about Mendelian genetics and Punnett squares, and that is the only part of my post you replied to? Youâre a kook.
Youâre also a totally wrong kook.
For diploid organisms, the alleles in the descents of the founders of a lineage will not all be from the founders, since mating outside the lineage can introduce other alleles. The frequency of a new allele in a diploid population will be 1/2n,not 1/n. Random walk calculations are not relevant to the creation of new alleles, only their subsequent spreading. Punnett squares are relevant - they provide the probabilities for the steps in the random walk.
Youâre also wrong that the limiting number of members in a diploid lineage is 2 (itâs 1); that allele lineages start with a single male and a single female (they start with individuals of either gender in larger populations); that the probability of a recombination event is computable from the allele frequencies (you also need to know the assortiveness, the zygosity and possibly which chromosome the alleles are on); that selection is competition; that competition slows adaptation; that E. coli and humans donât have a common ancestor; and that those people who know they do are ignoring Kishonyâs and Lenskiâs work. Those are just from one post.
Ask yours.
2 Likes
@kleinman at this point the entirety of your claims hinge on your understanding of genetics and mathematics being correct and the rest of the world being wrong. This is exceedingly unlikely. Youâve had the opportunity to make a case and been answered by people qualified in their respective areas. I think itâs time to bring the topic to a close, any closing thoughts?
2 Likes
The claims of the guy that thinks 90m/100m=90m hinges on his math being right? Uh oh.
1 Like
The reason that you are not seeing the problem is that you donât have a very good understanding of the physics and mathematics of evolution. It takes energy (food) to replicate, humans have much more food than chimps => human population >> chimp population. It is simple 1st law of thermodynamics example. Thatâs the competition part of evolution. Evolutionary adaptation is a second law of thermodynamics process. What a mutation does is disorder a genome. Biologists will call this diversification of the population but from a physical and mathematical point of view, this is disordering. The rate of this disordering can be computed with Markov chain mathematics if one writes out the equations correctly and you can compute the rate of one or more particular mutations occurring. Occassionally, in this disordering process, one or more mutations will give improved reproductive fitness for a particular environment. The only person who has posted in this thread who knows anything about Markov chain mathematics is Joe Felsenstein. He published what is called the âF81â model which is a derivative of the Jukes-Cantor model. You can read a little about it here:
The Jukes-Cantor and derivative models (including F81) are fundamentally flawed because they donât include population size as a variable in their models. My latest paper (in peer review at this time) shows how to correct this flaw. With this correction, the Markov model correctly simulates and predicts the DNA evolution of the Kishony experiment and other DNA evolutionary processes as well.
The Kishony Mega-Plate Experiment, a Markov Process
I donât expect you to understand this math, it takes some time and study to understand this but I would like to hear from Joe whether he thinks Iâm correct or not because Markov models are what phylogeneticists use to construct their phylogenetic trees.
In psychiatry, you really donât need to understand the laws of thermodynamics except when someone makes totally irrational claims about evolution that contradict those laws. That by definition is a loss of contact with reality. In my medical practice, I have had to treat drug-resistant infections many times so I thought it is worthwhile to understand this process a little better than was taught to me in my biology and genetics courses. Thatâs when I decided to put my engineering training into analyzing this process. A little study of the laws of thermodynamics works wonders in understanding evolution. Biologists might benefit from a better understanding of this subject (and a little more mathematics training wouldnât hurt either).
And you explain the differences between humans and chimpanzees by saying they just took different evolutionary trajectories from the common progenitor. Do you want to explain to us the physics and mathematics of DNA evolutionary trajectories? Letâs see if you can put some hard mathematical science to your claims. You might be able to help Joe Felsenstein with his work.
Is that so? Do you really believe that biologists have shown how humans and bacteria evolved in a common lineage? Those believers in that dogma canât even explain the simplest evolutionary experiments. And just because you donât have adequate training in physics and mathematics to understand my explanations of evolution doesnât mean I donât understand the subject. It means you donât have the skills to judge my explanations. Dan Eastman recognizes the validity of this math, at least for bacteria.
Now Dan needs to figure out that the mathematics of DNA evolution is the same for bacteria and any other replicator. You do have to take into account ploidy in that math and if you want to include recombination, that complicates the math a little. But DNA evolution is DNA evolution and when you understand this math, you will realize the human/bacteria common lineage idea represents a loss of contact with reality. I believe that problem falls under your purview.
Dan, you are not going to win this debate with semantics. Phenotype is the expression of the genotype.
Ask away.
Not at all. The mathematics for the DNA evolution of a beneficial, neutral, or detrimental mutation (allele) is identical. They all occur by a Markov chain random walk. Everytime a genome replicates, there is a probability that a mutation occurs at one or more sites in the genome. The frequency of occurrences of these mutations is given by the mutation rate. And each one of these mutations gives rise to a genotypic change that presents as a beneficial, neutral or detrimental phenotypic trait depending on the selection conditions of the environment. If that phenotypic trait is beneficial, it manifests by an increase in reproductive fitness for that new variant, if that phenotypic trait is neutral, it manifests by no change in reproductive fitness for that variant, and if that phenotypic trait is detrimental, it manifests by a decrease in reproductive fitness for that variant.
I know that. I said that for the benefit of those reading this thread that arenât. Every once in a while, one of my patients will tell me that they are going to Las Vegas or Reno to gamble. I tell them to give me all their money and that when they return, Iâll give them half their money back and they will be half ahead.
First of all, I am not saying that my papers are the forerunner for the downfall of evolution. You are conflating âevolutionâ with the âtheory of evolutionâ. My papers present the correct physics and mathematics of evolution. And since the mathematics of DNA evolution depends on probability theory, a subject which most biologists donât understand, it will require biologists that understand probability theory to turn this Titanic. Anyone that claims they understand evolution, I always ask this question. Does doubling the population size double the probability of a beneficial mutation occurring? If they say yes such as Dr. Swamidass and John Harshman did, I know they donât understand the mathematics of adaptive evolution.
There is another common line that will arise from time to time in these debates. People will claim that a series of microevolutionary changes add up to a macroevolutionary change. This is mathematically incorrect. Microevolutionary changes are random events, you compute the joint probability of random events using the multiplication rule. Thatâs why it takes a billion replications for each evolutionary step in the Kishony experiment (and every other DNA evolutionary adaption step). Itâs all about improving the probability of mutation B occurring on some member of the population that already has mutation A.
As far as the wager goes, I have no idea how long this will take to be accepted. I expect it will take years because the field of biology is filled with people with the mathematical skills similar to John Harshman who donât understand the difference between additive and complementary events.
Now, all you have to do is learn that DNA evolution in bacteria works no different than DNA evolution in any other replicator. Just multiply the number of member replications by the ploidy and you get the number of genome replications. Then, if you want to add recombination, first write out the mathematics of recombination correctly and superimpose those results on the mathematics of DNA evolution. Do it in a similar manner in which I worked out the mathematics for the Lenski experiment where I superimposed the mathematics of DNA evolutionary adaptation on the mathematics of a competition background (using a variation of the Haldane model).
I chose a journal with peer-reviewers that actually understand the math (unlike you), the correlation to the empirical examples used to derive this math, and understand the importance of this subject to the field of medicine. And Dan knows this. You and your ilk should expand your reading list.
Why not make it 3 billion possible beneficial mutations, no neutral, no detrimental mutations? Do you know that there may be 0 beneficial mutations? Do you know that there is no penicillin resistant group A strep has never been identified despite the fact that penicillin has been used for almost a century to treat these kinds of infections? You make a silly assumption to cover the fact that you donât know how to do the math if there is more than a single possible beneficial mutation. And to prove it, show us how the probabilities change if there are 2 possible beneficial mutations to a given selection condition. You wonât.
The probability of that recombination event occurring depends on the frequencies of the two variants with the beneficial alleles. So, in your 90,000,000 replications, you will get one beneficial A mutation allele and one B beneficial mutation allele. The rest of the population has neither mutation (allele). The probability of that recombination will be about 2*(1/90,000,000)(1/90,000,000). So, letâs say that as the generations go on, these beneficial alleles start to increase in numbers with respect to the rest of the population, but the frequencies of those variants will still be very small compared to the founder variants. What has to happen is that some selection pressure has to kill the founder variants so that the frequencies of the A and B variants increase. So, letâs say the frequency of the A variant is 0.5 and the frequency of the B variant is also 0.5. Then the probability of getting an A+B variant is 20.5*0.5=0.5. You have 4 possible recombinations, A+A, A+B, B+A, and B+B. This is exactly what is demonstrated with that yeast experiment. When the yeast population is not subject to selection, the founder alleles predominate in the population and the A+B recombination events are not observed. When the yeast population is put into the harsh environment, the founder variants are eliminated from the population so that the remaining yeasts are A and B variants at high frequency and the A+B recombination is observed.
But thatâs not the end of the story. How does the population take the next evolutionary step to improved fitness, the A2+B2 recombination event? You need DNA evolution to cause those A2 and B2 variants to appear, thatâs another 90,000,000 replications in order to take the next step improvement in fitness with recombination. You need to think more carefully about what it takes for recombination to achieve improvement in fitness. This is not a Hardy-Weinberg equilibrium problem.
Do you think the neutral varients just disappear? They donât unless selection is acting on the population such as in the yeast in a harsh environment example. And when the A+B variant does occur, it is the founder of a new lineage that needs DNA evolution (90,000,000 replications) to create the A2 and B2 alleles for the next possible A2+B2 recombination event. But the A2 and B2 variants will have a frequency of 1/90,000,000 each and unless selection kills the A+B (those members without the A2 or B2 mutation) variants, the A2+B2 recombination event will have a very low probability of occurring. You are just not understanding what has to happen to a population for these kinds of recombination events to have a reasonable probability of occurring. You have this vague idea that there are millions of beneficial alleles out there somehow recombining into a single lineage. You really need to learn how to put some mathematics to your vague ideas and correlate this math with real empirical examples.
Of course, I have, I even published the correct mathematics.
Random recombination and evolution of drug resistance
It happens that the mathematics for the DNA evolution for creating those alleles is the same for haploid and diploid replicators. The only difference is to get the correct number of genome replications, you have to multiply the member replications by the ploidy of the population replicating. The probability of a particular recombination event is proportional to the product of the frequencies of the particular alleles in the population. And it takes about 1/(mutation rate) replications to get the A and B beneficial variants the frequency of those A and B variants is about equal to the mutation rate for each.
Thatâs a fact Jack. Biologists still havenât written out correctly the mathematics for the probability of a particular recombination event occurring in a population. The best they can say is that there are lots of recombination events. Well, there are lots of mutations occurring all the time but that doesnât explain how particular beneficial mutations can accumulate in a lineage. The models biologists attempt to use to model DNA evolution are fundamentally incorrect. These models (based on Markov chain mathematics) donât include population size as a variable in their transition matrices. This gives totally inaccurate predictions when they use these models to calculate their phylogenetic trees and these models certainly donât predict the behavior of evolutionary experiments such as the Kishony and Lenski experiments. Iâve written a paper (presently in peer review) that corrects this.
The Kishony Mega-Plate Experiment, a Markov Process
I donât know how familiar you are with Markov chains but I know that Joe Felsenstein is, he has written one of these models. Joe has already been kind enough to correct Joshua on his inappropriate usage of neutral evolution to explain the genetic differences between humans and chimpanzees. I donât care whether that discussion continues on this topic or in a new topic but if Joe Felsenstein doesnât engage, most of the posters on this topic donât know the difference between additive and complimentary events let alone Markov chains, so it would be a total waste of time to discuss this topic without Joeâs participation.
We are talking about beneficial mutations. Donât try to change the subject.
If there are two beneficial mutations that happen in two different people those new variants will tend to increase in number over time. Do you agree or disagree?
As those new variants become more common in the population there is an increasing probability that carriers for each of the new variants will have offspring that carry both beneficial mutations. Do you agree or disagree?
If there are 50,000 possible beneficial mutations then you will have ~50,000 beneficial mutations in the population after 90 million replications. Do you agree or disagree?
1 Like
Tlhat would be the probability of at least one beneficial mutation occurring. A much better number to use would be the expected number of beneficial mutations, which would in fact double if the population size doubled. For similar reasons, population size is not relevant to neutral evolution.
Everyone should also notice that Kleinmanâs supposed models of natural selection do not model natural selection at all and in fact contain no parameter to account for the strength of selection.
Ah, but do they understand anything about evolution? What journal have you submitted your new paper to?
1 Like
Sure. The broad strokes of math is not that complicated thought, of course, the fine details can be more accurately described with more complicated math.
For any locus in the genome for which there are multiple alleles, the relative frequency of the various alleles will vary over time and generations.
Any given allele will ultimately eventually be completely eliminated form the population, or be fixed in every single member.
If the allele presents no selective advantage or disadvantage, this will be governed purely by genetic drift. IOW, pure, dumb luck.
If, OTOH, the allele presents the organism with disadvantages or advantages in terms of the likelihood of passing copies of its DNA on to future generations, then the likelihood of the allele being fixed eliminated will be adjusted accordingly.
Thatâs really all there is to it. Physics is largely irrelevant to understanding whatâs going on.
Hope that helps!
Yes. And not only in my opinion. Itâs pretty much the universal consensus among everyone here, as well as among the many, many other people who have encountered you over the years (decades?) on the internet.
2 Likes
Repeating claims we have previously refuted is pointless. It seems you simply have nothing more to say, otherwise you would have addressed these criticisms (mine and others) rather than ignoring them.
3 Likes
This topic was automatically closed 30 minutes after the last reply. New replies are no longer allowed.