Fred Williams' Mutation Rate Article Obsolete

quote:

---

In a nutshell, recombination increases the effectiveness of natural selection only in a harmful mutation environment. That is, only at some level of harmful mutation rate does sexual recombination start to become an advantage over clonal lineages.

---

Interesting interpretation. Especially when one considers the fact that you and many that use the same type of argument as you imply that all mutations are harmful to an extent. It seems logical that, therefore, sexual recombination is a benefit no matter what.Having read the papers in question, I fail to see the basis of your argument.

Presumably this stems from the same sort of phenomenon that holds up the belief of the anonymous scientist that Fred was talking about in the genetic information thread who said that nobody he knew who worked in genetics/ mol. biol. believed that the information in the genome was the result of evolution. They all just pretend they believe to get grant money.

quote:

---

Not true. Synergistic epistasis promotes some level of truncation selection.

---

I posted the quotation from the Nachman and Crowell article. It doesn't matter how Crow or anyone else use the terms; Nachman and Crowell explicitly distinguished between truncation selection and more general epistasis. They conclude that the latter is likely and that the former is unrealistic. You quoted them on the implausibility of truncation selection and also said that they concluded that it must occur. The truncation selection that they think is unrealistic is not anything they conclude must occur; your conflation of the two was dishonest.

quote:

---

In a nutshell, recombination increases the effectiveness of natural selection only in a harmful mutation environment. That is, only at some level of harmful mutation rate does sexual recombination start to become an advantage over clonal lineages.

---

The argument in your web page would be much more effective if it didn't miss the point completely. Recombination does not make a single beneficial mutation on a uniform genetic background fix faster; no one argues that it does. It does make multiple beneficial alleles fix faster. Without recombination, beneficial mutations have to occur in series, while with recombination they can occur in parallel, since different beneficial alleles can end up in the same offspring because of recombination. More generally, it will speed up the spread of beneficial alleles anytime there are multiple alleles undergoing (mild to moderate) selection at the same time, whether the selection is positive or negative. Since, as you like to point out, the deleterious mutation rate is high, this means recombination speeds up the spread of beneficial alleles in pretty much all naturally occurring populations.

quote:

---

The truncation selection that they think is unrealistic is not anything they conclude must occur; your conflation of the two was dishonest.

---

I see. So is Dr Crow also dishonest? You are completely mistaken. Nachman and Crowell comments on this were somewhat vague, which IMO is what led you to read into the text something that isn’t there. Apparently they are only distinguishing between *extreme* truncation selection and a lesser form of synergistic epistasis (which leads to a reduced level of truncation selection). If they believe as you do (which I very seriously doubt), then they can take it up with Dr Crow.Sfs, a while ago I decided to ignore/avoid evos who show a habitual difficulty telling the difference between willful errors (dishonesty) and unwillful errors (mistakes). I hope you are not one of them, so it’s up to you whether or not we conintue to debate. It will never cease to amaze me how quick many (not all) evolutionists will run toward the cry of dishonesty or liar when they are losing an argument. Do I think you are lying in your fallacious interpretation of Nachman/Crowell? No. I simply think you made a mistake. Do I think you are being dishonest with your simulation program that proved to have a serious flaw? No. BTW, you did not respond to by refutation of yoru simulation. Do you acknowledge the flaw?

quote:

---

It does make multiple beneficial alleles fix faster. Without recombination, beneficial mutations have to occur in series, while with recombination they can occur in parallel,

---

Parallel fixation isn't what it appears. Haldane showed each alelle still has to pay its own cost of fixation [1957, p 522]. Also, linkage disequilibrium is rarely found in sexually reproducing populations [Futuyma, Evolutionary Biology, 1998, p 300] so any impact via gene hitchhiking would be minimal.Also, asexual species aren't necessarily restricted to serial fixation only, considering the existence of plasmid transfer.

Why can't this be argued?

quote:

---

Recombination does not make a single beneficial mutation on a uniform genetic background fix faster; no one argues that it does

---

If this were true it would explain both postively Gould's "For all these reasons, I suspect that selection among deme-individuals holds an importance as yet unrealized ( and perhaps occurring in modes as yet unconceptualized) within our general picture of evolution. I have, in my carrer, witnessed three examples of widespread dismissal by ridicule as part of a professional ethos: the rejection of continental drift as physically inconceivable; the shunning of Goldschmidt's macroevolutionary ideas as danderous ot the Darwinian consensus; the the dismissal of group selection as addlepated nonsense (see""). Nothing in my intellectual life has made me feel more unconfortable."and negatively how Provine could have ignored my assertions of a twist in Wright's "general trial and error process" (NOT AN INFIINTE LAPLACIAN MIND SET) where frogs do not filter sound per say but essentially switch between longitudinal sound thru solid propagation and transverse waveforms. The science is all here. Lets do the work. I am thinking of using the physics of it to explain the different nervous system anatomy of salamanders vs anurans. I think Gould was simply disengenious when he drew attentin to Dobsanksy's attempt to figure Wright's landscape with mammmals. One needs really only try Croizat's Hylidae -MicroHylidae, Boa-Python or Agmadae-Iguanidae. The recent NATURE article which shows that more primative species stocks of Iguanids show more variation specifically indicates the likely chance to apply Age and Area to see if there is not an error in the CORE NS EXTENSION to an individual level the behvior of a lizard could remark. Somehow Provine felt that this infinity which is in the details of sound propagation was reason to distrust Creationists rather than the Fordian ecololgical influence in 1/2 his US collegues. I wish he had stuck to History. Reading Galelio on SoUND is all Gould's snake needs.1)USE Croizzt to fix the genetic background uniformaly (as to the geology) for the "world continent"2)Describe the incidence of the mutations in Muller's terms and background rate on this map3)Measure adaptation overseen by the GPS technology that sends back the data by TURNING Wright's Path ANALYSIS OPPOSITE to the direction the specimens ORIENT to4)adjust the way the Croizat track is standardized with a continually updating XML DTD.DO IT! Dont committ the DNA PRogramer instead to a hosiptial becasue unlike cloning which is up against nature DNA programming is only up against a check book. Lets add this to the list if we cant pony up Sharaf's horse or XenaS etc ... Spliting logic and language and logos wil not help. We need to come together. The other creatures ought care less but I do not speak for Galileo's Toad or the Bird's third TOE.

quote:

---

Sfs, even with ‘doTruncate = 0’ you are using SEVERE truncation selection!!!!

​

quote: /* Sort the offspring by their survival probability; take the top N */
for (ioff = 0; ioff < totoff; ioff++) {sortInd[ioff] = ioff;}

​

Bzzzt! You cannot sort offspring by survival probability, then lop off the bottom! This is classic truncation selection!

---

No, this wasn't truncation selection: a couple of lines earlier, the fitness (which is the real survival probability -- the code is commented incorrectly) was used to generate random survival. Even the most fit individual had a substantial probability of failing to reproduce, which is not the way truncation selection works.On the other hand, the code was also not correct: the trick I used got the selection coefficients wrong. So I rewrote it, doing the job properly. I now calculate the number of offspring with zero, one, two, etc deleterious mutations. I then draw the number of offspring in each class that survive to reproduce from a multinomial distribution, with the probability for each class being given by the number of offspring available times the fitness for that many mutations (which is just (1-s)^m, where s is the selection coefficient and m is the number of mutations). As expected, this model also reaches equilibrium, although at a higher mean number of deleterious alleles than the version that had the selection done incorrectly. The new code can be found at the link I posted earlier. The corrected results, for 1.6 deleterious mutations per generation, 3.0/gen, and 3.0 but with truncation selection applied, are:
.

quote:

---

I see. So is Dr Crow also dishonest? You are completely mistaken. Nachman and Crowell comments on this were somewhat vague, which IMO is what led you to read into the text something that isn’t there. Apparently they are only distinguishing between *extreme* truncation selection and a lesser form of synergistic epistasis (which leads to a reduced level of truncation selection). If they believe as you do (which I very seriously doubt), then they can take it up with Dr Crow.

---

As far as I know, Crow has never misrepresented their statements. You have. You said that they propose a kind of selection as the solution to the problem of deleterious alleles, and that they also think that that same kind of selection is unrealistic. That is, as I said before, dishonest. I have no idea whether you know your statement was dishonest, and I don't much care, but it was.

quote:

---

Parallel fixation isn't what it appears.

---

If you want to formulate an argument against recombination speeding fixation of beneficial alleles, by all means do so. The argument in your website isn't one, however, since it fails to address the actual mechanism proposed. You haven't done anything at all to support the claim you made, which was that recombination slows down the spread of beneficial alleles. (Nor have you provided any support for your claim that Crow's mechanism only slowed down degradation.)

quote:

---

Haldane showed each alelle still has to pay its own cost of fixation [1957, p 522].

---

For mutations that compensate for a steadily degrading environment, as I recall. In your argument you'd better demonstrate that this model has any relevance for most real-world cases of evolution.[This message has been edited by sfs, 10-19-2003]

quote:

---

If you want to formulate an argument against recombination speeding fixation of beneficial alleles, by all means do so.

---

Also, I might refer Fred to here - I would appreciate if he could explain why good genes are fixating there, when it's not using truncation selection, and harmful mutations are predominant.------------------
"Illuminant light,
illuminate me."

I just have a wee bit of spare time to answer a few posts

quote:

---

If you want to formulate an argument against recombination speeding fixation of beneficial alleles, by all means do so.

---

I thought I already did, but as is typically the case with evolutionists I guess you didn’t like the answer.Only half the offspring will receive the beneficial allele. Out of this immediate 50% barrier, only a percentage of them will reproduce to the next generation. Such a barrier doesn’t exist in clonal lineages.To make matters worse, most traits are polygenic, or quantitative traits. So beneficial mutations can get buried further into the probability pile. For example, given a 5-gene trait where one of the genes is mutated that caused the increase in fitness, only 1 out of every 2^5 = 32 offspring would receive the same combination that produced the fitness increase.

quote:

---

For mutations that compensate for a steadily degrading environment, as I recall.

---

Baloney. Why don’t you get the paper instead of just recalling stuff you probably read second-hand on the internet.PS. I don’t have time to correct your newly-modified program again. You used severe truncation selection and now you are trying to deny it with excuses. To expose the complete folly of your program, why don’t you post a chart with 100 mutations/organism/generation for everyone to see. This would require 2e^-100 offspring to maintain equilibrium, which is a number that is larger than half of the estimated atoms in the universe! Yet your program will show a nice little smoothing out at the top, just in time to save humanity! Oh my!PS to Percy. I wanted to answer your comments about being anonymous but I don’t have time to find the thread. You misunderstood me. In general I don’t object to posters being anonymous, in fact I could care less. I only muse about those who resort to ad homenims while under anonymous pseudonyms. They don’t seem to realize that it only serves to hurt their credibility.

quote:

---

Also, I might refer Fred to here - I would appreciate if he could explain why good genes are fixating there, when it's not using truncation selection, and harmful mutations are predominant.

---

OK, I had time to look through this (the only section I did not check is the last portion that calculates the statistics). From what I can tell, it’s not a bad little program! Your selection method is reasonable, particularly since random deaths are somewhat accounted for (but not completely, but that’s another debate; see my program for a more realistic approach: 404 Not Found ). The bad news for you is that if you take what are well-known facts and apply them to your program, then you have no excuse whatsoever to believe evolution can occur. Your own program convicts you! More on that later.First, before looking at your program I noticed your premise is a strawman. I don’t know of any creation scientist with experience in population genetics or mathematics who denies that beneficial alleles can fixate (or increases overall fitness) in an environment under selection where most mutations are harmful. What primarily matters is the ratio of beneficial mutations to deleterious mutations, and that realistic selection is occurring. You can make many things happen when you tinker with the mutation ratio and selection. In your example you tinker mightily with the mutation ratio to create your fantasyland.So let’s consider your example (aka fantasyland), that produces an increasing fitness average across the population. Well, for starters you selected a beneficial mutation rate of 1 out of every 12 (8%)!!! A whopping 1 out of 25 have a 10% advantage! No wonder your program succeeded![NOTE: You meant to set your beneficial rate to 4%, but there appears to be a bug in your program. You intended 4% *total* beneficial, with .2% very beneficial at 10%, but your numbers only added up to 96%. Because your last ‘if’ statement doesn’t protect against this, it will force your 10% advantage bracket to capture the last 4%, giving a total of 4.2% in the 10% advantageous bracket.]But even with a corrected 4% as opposed to 8% beneficial, I would not be surprised to see a 1 in 25 beneficial clip still move the fitness average upward. This is in a nutshell the religion of evolution: faith in chance. Darwinian mutations are *mistakes*, or random copy errors. No one believes the frequency of beneficial mutations is 1 in 1000, let alone your fantasyland 1 in 12! Is 1 in a million possible? Unlikely. It certainly isn’t supported by the years of research on bacteria and drosophila where we can only find a handful of questionable examples. But if one is to believe in evolution, one must believe in the impossible, ie an impossible ratio of beneficial chance mistakes.To make matters worse, your deleterious mutation rate is probably too low, and an unrealistically high number of deleterious mutations are removed each generation. Your gene mutation rate is set to 1 in 50. That means out of 100000 genes, roughly 2000 will contain a mutation of some sort. At least 110 of the 120 (2000*.06) lethal mutations will be removed. For simplicity we’ll favorably assume exactly half of the remaining 720 (2000*.36) deleterious mutations will be removed (the vast majority, or 98% of comparisons will be against genes with a fitness of 1). So after one generation, at most 370 deleterious mutations will remain. This means at best only 1 in three organisms will have a deleterious mutation going into the next generation (a poisson distribution will lower the ratio somewhat, as some organisms will have multiple mutations). This low mutation rate is yet another item that is not supported by scientific evidence. The lowest deleterious rate I’ve seen from the scientific community is 1.6 per organism/generation (after selection), yet yours is roughly 0.3 per organism.Now time to put the fantasy on its head. Let’s now apply some real world numbers, yet still be extremely favorable to your cause. Let’s be generous and set the beneficial mutation rate to an extremely favorable 1 in 1000 (still fantasyland, but not as much as your 1 in 12). Go ahead and leave the overall mutation rate to an unrealistic 1 in 50, and deleterious rate at 42% (yielding roughly .3/organism). Now run your program and tell us the results! Run it as many times as you possibly can! You know and I know and everyone else here knows you are against the insurmountable barrier of chance. You will always lose, always. Your program will show a steady fitness decline. ALWAYS!Your own program should be enough to convict you that evolution is a fairytale, even given extremely favorable and downright unrealistic conditions! However, since evolution is a religion and not science, these truths will have no impact. Question of the day: Why is it you can still cling to a belief system that your own program proves is an impossibility?[PS. I have very little time to post these days as a new assigment at work will keep me busy for some time, so y'all will have to be patient for replies]

The fact that each child has two parents cancels out the factor of 0.5The way you are treating polygenic traits requires some justification, too.How do you justify the assumption that there is no benefit unless the correct alleles for all 5 genes are in place ?How do you justify the assumption that the other parent has none of the 5 correct alleles ?

(My apology for the delay in posting... I had to run to do a radio show, and then had to wait for the now incredibly slow program to run. Update, 4:32 PM... grr, this is taking way too long, let's recompile with heavy optimization, and move it to a 2.8 Ghz system and restart... ah, this is going much faster, although it's still painfully slow... Update, 5:00 PM... it's seems to be about 25 generations per hour. I'll post at 5:30 PM, but let it keep running, and update the output file at regular intervals (linked from this post). Update, 5:30: Let's go!).(Edit: Post moved to http://EvC Forum: Fixation of genetics - program where it belongs)[This message has been edited by Rei, 10-24-2003]

Hi Fred,I've posted a reply at Message 16 of the Fixation of genetics - program thread.--Percy

I gave your program a try, Fred, but:

• The header file std_defs.h isn't available in the program directory, but I was able to create one.
• There's a syntax error at line 18 of header file prog_defs.h
• You omitted the required semicolon after the definitions of structures ORGANISM_Q and MUTATION_Q.
• You reference structure ORGANISM at line 35 of header file prog_def.h before you define it causing a syntax error.
• Same for structure MUTATION_Q at line 48.
• You need prototypes for routines get_parms(), init_parms(), mate_organisms(), display_results(), program_break() and so forth if you're going to reference them before their definition.
• You reference many undeclared variables, too many to enumerate.
• Routine display_results() in main.cpp is not defined.

Your program is filled with syntactical and semantic errors, Fred - this program never ran. If your program were presented at a review meeting everyone would laugh and then call personnel. If you fix these problems I'll give it another try.--Percy[This message has been edited by Percipient, 10-26-2003]