Rebuttal To Creationists - "Since We Can't Directly Observe Evolution..."

Kleinman:

---

Idiot, your frequencies are not mutually exclusive.

Taq:

---

Then the addition rule does not apply.

---

---

It does dumb, dumb. You simply have to subtract off the intersection so that you don't count those members twice. 26 years of research and you never learned this simple mathematical rule.

Kleinman:

---

Descent with modification is modeled using a Markov chain and Markov chains are entropy producing processes.

Taq:

---

Let's use two different DNA sequences:
​
TTTCTATAGTCCTTGAGAGGAGGAGTCGTC
TTTCTAGAGTCCTTGAGAGGAGGAGTCGTC
​
They differ by a G:T early in the sequence. Which of these has lower entropy, and why?

---

---

It's your example, go for it. I'd rather do Markov chains for real evolutionary examples such as the Kishony experiment. Why don't you do that example because we can compare it with a real evolutionary example, not some made-up example? But go ahead, make up a story.

Kleinman:

---

5e8 replications per day, times 365 days, times 30 years and about 100 adaptive mutations in that time.

Taq:

---

Those are just the adaptive mutations that fixed. There is no reason to think that these are all the adaptive mutations that occurred. Even your own discussions claim that adaptive mutations will occur and be driven to extinction by other adaptive mutations. Even more, adaptive mutations can be masked by deleterious mutations.
​
Apparently, you don't understand how evolution works in the Lenski experiment. You certainly don't understand how evolution works in the Lederberg experiment where beneficial mutations for phage resistance occur 1 in every 10 million replications. Nor do you understand that the rate of adaptive mutations in one experiment can not be applied to another, much less to all species in all environments.

---

---

They didn't fix because they were driven to extinction. There is a cost to natural selection. And make your case for UCD with phages.

Kleinman:

---

So you don't know how to do a Markov chain.

Taq:

---

Your attempts to change the subject are quite informative. Why can't you answer this question?
​
​
TTTCTATAGTCCTTGAGAGGAGGAGTCGTC
TTTCTAGAGTCCTTGAGAGGAGGAGTCGTC
​
They differ by a G:T early in the sequence. Which of these has lower entropy, and why?

---

---

I do Markov chains for real examples so I can compare experimental data with mathematical results. You make up examples so you can tell a story. You don't know how to do a Markov calculation. Here's how you do a Markov calculation for a real example:
The Kishony Mega-Plate Experiment, a Markov Process

Kleinman:

---

Sure it does idiot, just subtract off the intersection of the subsets.

Taq:

---

Then this means the two variants can be any combination of frequencies, correct?
​
If so, doesn't this defeat the very argument you were making based on the addition rule?

---

---

The point of this calculation is to produce an AB variant. If the frequencies of A and B are very low in the population, then the probability of that happening by recombination will be very low. Are you trying to produce a variant with cystic fibrosis and achondroplasia? If you are, you had better correct your math.

Kleinman:

---

Oh, I see, there were millions of adaptive mutations, they just didn't fix.

Taq:

---

Isn't that what you have been arguing the whole time? You have been arguing that clonal interference occurs which can only drive one adaptive mutation to fixation which driving lesser adaptive mutations to extinction.
​
And you still don't have an explanation for the Lederberg experiment.

---

---

Son of a gun, and all along I thought we had lost them. Isn't that how fixation works, the most fit variant fixes in the population while the less fit variants are driven to extinction.

Kleinman:

---

When you brought up Markov chains, I suggested that you do a real example, the Kishony experiment. Then you plop down some speculative example. You can't do a Markov chain of a real example.

Taq:

---

If you can't find the relative difference in entropy between two sequences then what hope do you have of doing the same for the multiple sequences produced by a Markov process?

---

---

I'll stick with doing Markov calculations with real processes. I'll let you make up examples that only exist in your head. When are you going to tell us a story about those two sequences?

Kleinman:

---

That's what happens, and that's what happened in the Lenski experiment.

Taq:

---

Then by your own admission there would have been adaptive mutations that didn't fix. Therefore, you can't use the fixation rate to determine the beneficial mutation rate.

---

---

Of course, they didn't fix because they were not the most fit variant.

Kleinman:

---

You explain the Lederberg experiment if you think it proves UCD.

Taq:

---

I never said that the Lederberg experiment proves UCD. What are you talking about?
​
Can you or can you not explain the results of the Lederberg experiment? This is a simple case of evolution in E. coli to a single environmental challenge. Can you explain it?

---

---

Oh, I see. Then what is the reason that you bring up phages? It has nothing to do with UCD, so what is your point?

Kleinman:

---

You don't know how to do a Markov calculation.

Taq:

---

You don't know how to do measurements of entropy.

---

---

Sure I do. It just isn't necessary to calculate entropy to do a Markov calculation of a real experiment. What is important is to know how a population diversifies and you do that by writing a Markov calculation for the evolutionary process. You can calculate the probability of an adaptive mutation occurring.

Kleinman:

---

Foundering is what you do when I ask you simple questions about ERVs.

Taq:

---

That's strange. I gave you a pretty decent rundown of the ERV evidence and you didn't have a response.

---

---

You can cherry-pick data all you want, it doesn't impress me. Anyone that thinks that 8% of the human genome consists of viral DNA has no idea what a virus does to a cell.

Kleinman:

---

Sure I do, but you don't need to.

Taq:

---

You need to.
​
TTTCTATAGTCCTTGAGAGGAGGAGTCGTC
TTTCTAGAGTCCTTGAGAGGAGGAGTCGTC
​
They differ by a G:T early in the sequence. Which of these has lower entropy, and why?

---

---

No, you need to in order to make up a story. I'll stick with Markov processes with real examples such as the Kishony experiment. This is how you compute the diversification of a population.

Kleinman:

---

You only need to be able to compute the occurrence of a single adaptive mutation in a single selection pressure environment or two adaptive mutations in a two selection pressure environment, or three adaptive mutations in a three selection pressure environment,...

Taq:

---

Let's take a look at the sickle cell trait. In regions with endemic malaria this is an adaptive mutation. However, in regions without malaria the side effects of sickle cell anemia make the mutation non-adaptive. So how is the same mutation an increase in entropy in one environment and a decrease in entropy in another?

---

---

The selection conditions are determined by the environment.

Kleinman:

---

Nope, that's all I'm assuming.

PaulK:

---

Well it seems very confused. Aside from mixing up alleles and genomes you need some explanation of how you can have C at both loci. A recent gene duplication?

---

---

C is any allele that is not A or B. C doesn't have to be all the same allele, they just have to be different than A and B.

Kleinman:

---

Nope, I only assume that whatever gene is at a given locus, is homogenous. You assume they are not homogenous.

PaulK:

---

I certainly did not make any such assumption. But just to make you happy I will write it out again with homozygosity explicitly assumed,
​
If one locus has alleles A and C (and no others) and the other has B and D and all individuals are homozygous, the frequencies of AA and CC must sum to 1. Adding the frequency of BB is pointless. (Indeed, if no individual has AA and BB the overlap of BB and CC must be every individual with BB).

---

---

Sure you do. You do that when you say one locus has A and C alleles and a second locus has B and D alleles. That makes the loci heterogeneous. I'm saying that one locus is AA or CC and the other locus is BB or CC. Then if an AA mates with a BB, you will always get an AB offspring, ie., no Mendelian Genetics. The recombination event is guaranteed to work.

quote:

---

C is any allele that is not A or B. C doesn't have to be all the same allele, they just have to be different than A and B.

---

Then I suggest that you use less obfuscatory terminology.

quote:

---

Sure you do. You do that when you say one locus has A and C alleles and a second locus has B and D alleles.

---

In the same way that you “assume heterozygosity” when you say that the first locus can have A or C. I.e not at all.And it should be perfectly clear that nothing in the following text supported your misinterpretation at all. It was just a straightforward explanation of the rule of 1.

Kleinman:

---

C is any allele that is not A or B. C doesn't have to be all the same allele, they just have to be different than A and B.

PaulK:

---

Then I suggest that you use less obfuscatory terminology.

---

---

I'm sorry that this simple concept confuses you. I wrote the following in the published paper: It couldn't be simpler. Try careful reading.

Kleinman:

---

Sure you do. You do that when you say one locus has A and C alleles and a second locus has B and D alleles.

PaulK:

---

In the same way that you “assume heterozygosity” when you say that the first locus can have A or C. I.e not at all.
​
And it should be perfectly clear that nothing in the following text supported your misinterpretation at all. It was just a straightforward explanation of the rule of 1.

---

---

I'm not misinterpreting anything. You just don't read very carefully. That's why you screw up such a simple concept.

quote:

---

I'm sorry that this simple concept confuses you. I wrote the following in the published paper

---

Since the explanation is rather critical to understanding it, leaving it out is your failure,

quote:

---

I'm not misinterpreting anything

---

So you really do think that the rule of 1 only applies if we assume heterozygosity.

quote:

---

You just don't read very carefully

---

Oh, I get it. This is where you attribute your faults to others. You didn’t read carefully and misunderstood so you have to pretend that the writer didn’t read carefully.

quote:

---

That's why you screw up such a simple concept.

---

I’m not the one that screwed up. But again, you have to attribute your faults to others.