Understanding through Discussion


Welcome! You are not logged in. [ Login ]
EvC Forum active members: 85 (8915 total)
Current session began: 
Page Loaded: 07-22-2019 6:39 AM
21 online now:
(21 visitors)
Chatting now:  Chat room empty
Newest Member: 4petdinos
Upcoming Birthdays: anglagard
Post Volume:
Total: 857,297 Year: 12,333/19,786 Month: 2,114/2,641 Week: 69/554 Day: 6/63 Hour: 0/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Can you disprove this secular argument against evolution?
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(2)
Message 139 of 293 (804406)
04-09-2017 8:45 AM
Reply to: Message 1 by forexhr
03-31-2017 3:26 AM


The Texas sharpshooter rides again
forexhr writes:

But that begs the question: how did this selectable combination of CHNOPS(protein) came to be? This is the crucial and the most important question. There are virtually infinite number of ways in which CHNOPS comprising protein can be arranged, and most are junk, or non-selectable arrangements. For e.g. for a protein 92 AA long, with 10e122 possible AA combinatios, there is only 1 in every 10e63 functional sequence*. On the other hand, published extreme upper limit estimates puts the maximum number of mutations or CHNOPS re-arrangements at 10e43**. So, the total number of evolutionary CHNOPS re-arrangements is 20 orders of magnitude insufficient to find only one selectable state for evolution to preserve - a protein, let alone molecular machines, organs or organ systems.

*Functionally acceptable substitutions in two alpha-helical regions of lambda repressor. Reidhaar-Olson JF, Sauer RT.

**How much of protein sequence space has been explored by life on Earth?, David T.F Dryden, Andrew R Thomson, John H White

I've looked at the two papers you've linked to, and I'll try to explain what how you've come to the wrong conclusion from them.

Supposing we have a random six-figure number generator. Certain numbers are special, let's say those that begin and end with "6", so we know that, given enough trials, about 1% of our numbers will be in that set. We do a trial, and by chance we get a special number, 673916. The probability of getting a special was 1/100, but the probability of that specific number was 1/1,000,000. If we viewed 673916 as a specific target, we'd consider ourselves amazingly lucky to hit it, but if just any special number is the target, then lucky, yes, but nothing amazing about it.

In that example, we know that 1% of numbers are special, but what's missing from your O.P. is the proportion of all random AA combinations that are special (functional in any way). One specific protein with many functional variations for a specific function totalling 1 in 10^63 of all random combinations doesn't tell us the proportion of combinations that are functional in some way. So, lets look at this paper: Functional Proteins from a random sequence library which comes up with an estimate of 1 in 10^11 combinations having some function. Then, for simplification, let's assume that all individual functional proteins with their variations would each only be ~1 in 10^63 of total combinations, like the one in your paper. Therefore, if a life system had run 10^43 random trials, and achieved a "hit" on 1 in 10^11, it would easily have have enough functional proteins for millions of species with billions of functions yet any and every specific protein would appear highly improbable in isolation, in a similar way that the number 673916 seems amazing in my number game if we consider it as a target.

So, there's nothing to disprove in your argument, because you've made a mistake in your way of viewing probabilities. The specific protein in the paper that gives you the 1 in 10^63 figure was never a target; no specific protein, organ or organism is essential to a life system, and all should look highly improbable individually if a model like the one I've described above is correct.

Edited by bluegenes, : missing word


This message is a reply to:
 Message 1 by forexhr, posted 03-31-2017 3:26 AM forexhr has responded

Replies to this message:
 Message 144 by forexhr, posted 04-09-2017 10:55 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


Message 169 of 293 (804785)
04-13-2017 4:31 AM
Reply to: Message 144 by forexhr
04-09-2017 10:55 AM


Re: The Texas sharpshooter rides again
forexhr writes:

You made one critical error in your reasoning: you defined function with relation to - nothing. In biology, function is always defined with relation to something. Even you paper has done this, I quote: "we selected functional proteins by enriching for those that bind to ATP".

There's no error. "Function" is used in biology in both a general and specific way. The Szostak paper (Functional Proteins from a random sequence library) I linked to does both, and the conclusion is general.

quote:
In conclusion, we suggest that functional proteins are sufficiently common in protein sequence space (roughly 1 in 10^11) that they may be discovered by entirely stochastic means, such as presumably operated when proteins were first used by living organisms.

Surely you understood the general uses in the second paper you linked to in the O.P. here, and the one on functional DNA you linked to in your second post here?.

forexhr writes:

In my paper, this function was lambda repressor fold. If the ability to bind to ATP is considered a closed stuctural niche, only for that function you need to spend 10^11 evolutionary resources. If to this we add resources needed to extract lambda repressor fold, we are already 31 orders of magnitude short with regards to all evolutionary resources. Now add to this all differnt protein folds, organs and morphological structures and you will get a clear picture of the extent of the problem
Also, in your paper they started with small proteins, each containing 80 amino acids. On the other hand, average length of, for e.g. human protein, is 480 amino acids (1). Finally, your random six-figure number generator is totally unrelated to biology since its sequence space is 1.000.000 while for an average protein this space is 10^624 in size.

My numbers were to illustrate your probability mistakes.

For example, you are mistaking the resources required to hit all possible functional 1 in 10^624 AA combinations for the resources required to get any. If there are 10^523 such functional proteins, then 1 in 10^11 searches (random assemblies of 480AAs ) would hit on one. 10^17 searches would provide 1,000,000, and 10^20 searches would hit on 1 billion of them, already plenty for a very diverse life system.

The same applies for your 1 in 10^63 estimate for a related group of proteins with a specific function. If there are 10^52 such groups with some potential function, then 1 in 10^11 searches would hit on one such. 10^17 searches would provide 1,000,000, and 10^20 would hit on 1 billion of them. The Lambda phage doesn't have to exist at this point or any other, neither does its repressor, and neither does the E-coli host. Life has no specific targets. But pick out any single protein in the system and it will appear to be remotely unlikely to those who don't understand probabilities.

If you can learn to understand the probabilities, then there are some technical points that are interesting, as well. It's not hard to find completely unrelated proteins that can perform the same function, for example, something that might support the conclusion of the second paper you linked to.

Edited by bluegenes, : No reason given.

Edited by bluegenes, : extra word


This message is a reply to:
 Message 144 by forexhr, posted 04-09-2017 10:55 AM forexhr has responded

Replies to this message:
 Message 170 by forexhr, posted 04-13-2017 7:06 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(1)
Message 216 of 293 (805093)
04-15-2017 12:56 PM
Reply to: Message 170 by forexhr
04-13-2017 7:06 AM


Re: The Texas sharpshooter rides again
.
forexhr writes:

What does functional protein that binds to ATP has to do with the development of all specific bio-functions that we observe at all levels of bio-organization.

Well, you like targets, so we can think up an experiment together. We are asking ourselves the general question: if a biological species was randomly shuffling around amino acids via mutations on its genome, about how many attempts might it take to come up with a new protein that is useful to it? So, we could pick out a random specific target, and test a whole bunch of random AA sequences for it. We don't have to test our AA sequences out for any other of the many specific functions they might perform in order to infer that our result gives an approximation of the general answer to our question.

Functional Proteins from a random sequence library

Fortunately, Szostak and Co. have done it for us. It's ~1 in 10^11.

forexhr writes:

If we consider ATP binding function as "general" what would this generality have to do with this specific structural niches: intron-exon gene structure, female reproductive system, lactose, ability of the lambda repressor to regulate expresion of cI protein? Can you fulfil these niches with ATP binding function? No. Can you fulfil the 'contraction' niche of striated muscle tissues, with this protein composed of 80 amino acids? No. You need a specific protein protein composed of ~27,000 to ~33,000 amino acids that is called Titin. So the Szostak paper demostrated that there has been enough sampling to cover the entire functional landscape of ATP binding protein. But what follows from that in the context of above mentioned niches? Absolutely nothing.

The Szostak sample wasn't tested for other functions. It hit its target, and would have hit many others. It tells you a lot. Amongst other things, because four completely unrelated proteins that had some ATP binding function were found in the sample, you should no longer be using the 1 in 10^63 figure (from the first paper linked to in the O.P.) in the way that you are, because it's a very rough estimate on the proportion of search space that could be derived from variations on the current Lamda protein, and doesn't include any completely unrelated ones that could cause the function, although the authors do indirectly touch on this possibility in the paper.

Also, when you talk about protein size, you need to understand degeneracy in proteins. As the protein increases in size, the number of functional variants increases exponentially, so if we could derive, for example, 10^50 functional variants on one of the 80AA ATP binding proteins found in the Szostak experiment, we'd expect ~10^100 functionals on a 180AA protein.

forexhr writes:

You committed the same mistake again. Let me use this example for illustration. If you have 1000 generally meaningful 10 letter words, whose sequence space is 26^10=141167095653376, and if you have enough resources(trials) to explore this sequence sapce, for example 150000000000000 trials, than by your logic, 150000000000000 trials would stumble across all 1000 generally meaningful 10 letter words.But here is the problem. How would you define what is meaningful in a specific context? In a language, meaningful is defined by other words in a sentence. For example: this 10 letter word - "authorized" is meaningful in the context of this sentence: The defender admits that he authorized all operations, but this 10 letter word - "birthplace" is not. Since you don't know in advance what sentence or context will appear, it is impossible to "select" a specific meaningful 10 letter word if you stumble across it during a random search.

I was assuming context. What I was saying, using your analogy, was more like if about 1412 combinations are potentially functional (in this context) then 10^11 searches would find one, and that the same principle would apply as an approximation for all contexts (in general).

One of the reasons it doesn't appear to work in your analogy but does for function in biology is that life is a terrible speller, so there would be about 10^20 ways of spelling a 30 letter word (according to the paper in your O.P.) and about 10^6 ways of spelling "authorized" instead of the two we have in modern English. So, we actually would hit on authorized itself fairly easily, inside Szostak's estimate, without depending on ten letter synonyms. So, you could get an average length 100,000 word/protein novel with 10^16 "resources", and you could have 1,000,000,000,000,000,000,000,000,000 novels if all the search resources available to life so far (according to your second O.P. paper) had been available.

The first paper you linked to in the O.P. covers the different "spellings", but not the different "synonyms" and all their different spellings.

The Szostak experiment didn't test its sample for all functions, only one. The thinking goes, "if the organism could benefit from function x, then about how many random arrangements of AAs might it take to get it, so let's test random arrangements for an x." They could have chosen other specific functions, and the same thing could be tried for the Lamda repressor to give a much more accurate picture of the proportion of proteins of all families that would give some minimal selectable function.

forexhr writes:

This is even more obvious in the context of biology. Proteins are meaningful only in the context of currently opened structural or environmental niches.

If we want to add just one completely new functional protein to our life system today, how many such niches are available to all species? One? An uncountable number? How many per. species?

forexhr writes:

For example if this niche arises "The defender admits that he .......... all operations", then to fulfil the niche you need to search for the word "authorized". And to find this word you need to spend almost all of your resources since sequence space of 10 letter words is 26^10=141167095653376. If another niche arises where the word "birthplace" is meaningful, then again, you need 150000000000000 resources to find it. But you alredy spend all your resources for the previous search and you cannot proceed.

You could have written:

For example if this niche arises "ATP binding" then to fulfil the niche you need to search for a protein that binds ATP, and to find this you need a fraction of your resources (10^11 out of 10^43). If another niche…..

But that would be real biology, wouldn't it?

You still seem to be reading your novel backwards and assuming there can only be one novel and that life systems have a plot with an ending.

Once again, the resources for "authorized" in biology would be a lot less because there would be about 10^6 functional spellings on a ten letter stretch. A 30 letter word would have about 10^20 spellings, according to the first paper in your O.P.

forexhr writes:

To conclude, in your reasoning you presupposed that evolution has a foresight and knows what structural or environmental niches will emerge in the future, so when the random search stumbles across the protein that is meaningful in the context of these future niches, evolution would simply select it, then put it aside and wait for niches to emerge in some random point in the future. In short, you presupposed that evolution has the supernatural powers.

Not at all. A species could have many possible structural or environmental niches open to it at any one time, and these can be available for long periods of time. Instead of "authorized" or one of its synonyms, you get "designed" or "loved" or "hated" and the story goes off in a different direction, as it would if you substituted "denied" for "admits" and so on.

It's because you like targets that I dug up the Szostak paper for you. It has one, so I thought you'd like it....


This message is a reply to:
 Message 170 by forexhr, posted 04-13-2017 7:06 AM forexhr has responded

Replies to this message:
 Message 220 by forexhr, posted 04-17-2017 6:34 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(3)
Message 218 of 293 (805137)
04-16-2017 8:21 AM
Reply to: Message 170 by forexhr
04-13-2017 7:06 AM


Lambda hits Target with spectacular speed.
forexhr writes:

Since you don't know in advance what sentence or context will appear, it is impossible to "select" a specific meaningful 10 letter word if you stumble across it during a random search.

This is even more obvious in the context of biology. Proteins are meaningful only in the context of currently opened structural or environmental niches.

Amazing how quickly the impossible can be achieved. The bacterial phage of the O.P. paper, Lambda, hits a target for forexhr.

Lambda achieves the impossible in about 8 days!

The new function often arrives between 8 and 20 days, The whole paper's well worth a read, but here's an extract for the lazy.

quote:

There are four striking cases of parallel evolution of the J protein in the phage that target OmpF. In two cases, the mutations were identical across all 24 populations, while in two others there were slight variations (Fig. 3). In particular, all J alleles from phage able to infect through OmpF had the A-to-G mutation at nucleotide position 3034 and G-to-A mutation at position 3319. Also, all of them had a mutation at either position 3320 or 3321, affecting the same codon (amino-acid residue 1107) as the mutation at position 3319. Finally, all J alleles had at least one mutation between positions 2969 and 2999 (amino-acid residues 990 to 1000).

Each of these mutations or classes of mutation was also found in at least one of the phage that retained the ancestral host-range, although none of them had all four together (Fig. 3). Two LamB-dependent isolates, F2 and H4, had three of the mutations, as did EvoA from the initial experiment (Fig. 2), yet none produced clearing on lawns of lamB mutants.

The correspondence between the use of the OmpF receptor and the presence of these four mutations, coupled with the observation that phage having only three of the four cannot use OmpF, provides evidence that all four are required for λ to infect through OmpF. We performed two additional assays to confirm that only phage with all four mutations can infect lamB mutants (16). The assays were performed using isolates EvoA, F2, and H4 that each had three of the four canonical mutations and D7 that had all four and no others. Only D7 exhibited a measureable adsorption rate on lamB mutant cells (fig. S5), and it was also the only one that reproduced on lamB mutants in the medium used in the evolution experiments (fig. S6). These findings indicate an “all-or-none” form of epistasis among the four mutations responsible for the novel receptor phenotype.


All four mutations are required.

As I said before forexhr, I know you like targets, so here's your favourite organism hitting a target with a new four letter word.


This message is a reply to:
 Message 170 by forexhr, posted 04-13-2017 7:06 AM forexhr has responded

Replies to this message:
 Message 221 by forexhr, posted 04-17-2017 6:47 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(1)
Message 244 of 293 (805448)
04-18-2017 3:08 PM
Reply to: Message 220 by forexhr
04-17-2017 6:34 AM


Re: The Texas sharpshooter rides again
forexhr writes:

Isn't it interesting how your entire response is focusing on issues that we both agree with.

Entire? If you agree with my post in its entirety, then you are disagreeing with your own O.P.

forexhr writes:

I agree that the Szostak sample wasn't tested for other functions and that it hit its target. I agree that as the polypeptide chain grows in size, the number of functional combinatorial possibilities increases exponentially. I agree that there are many functional amino acid sequences with respect to the vast combinatorial space of possibilities - I stated this explicitly few times already - this is from my last post: "Given the study referenced in the O.P., there are 10^56 "favorable outcomes" and 10^119 possible outcomes".

Right in everything except the numbers. In terms of the repressor function, there are probably far more "favourable outcomes". There could well be ~10^108 (Szostak). That's because the paper looks at function in a "family" of proteins derived from the one that's actually there, but not at the sequentially "unrelated" proteins that also might perform that specific function. The 10^57 would be identifiable as "family" by sequence similarity - they would all have at least 20 identical positions (there were 8 that they considered unchangeable in the sample of 33). That's one of your technical errors. I don't know why you made it, because you must know that there are "super-families" identifiable by function but not by sequence, and the paper doesn't actually claim that their 1 in 10^63 figure covers all proteins that could give a minimal selectable function in that context. As I said in my previous post, the first paper in your O.P. covers the highly degenerate "spellings" of the protein that's there, but not its "synonyms" and their multiple spellings.

forexhr writes:

So, you spent your entire post to repeat something that has already been said and that nobody disagrees with. But, you din't even consider to critically examine the logical consequences of the paper you talked about - if 10^11, out of 10^43 evolutionary resources must be spent just to extract one simple binding function, then how much resources must be spent to extract complex and structurally independent functionas like: conformational complementarity of enzyme and substrate, fertilisation, editing of the nascent precursor messenger RNA transcript, synchronization of the kicks of the hopping insect's legs via gear mechanism, blood pumping, visual perception,.... According to Wikipedia, whose editors savage anyone who criticizes the theory of evolution, "... Complex, image-forming eyes have evolved independently some 50 to 100 times."

Why not just skip to an entire complex organism, like Homo Sapiens. If you "critically examine the logical consequences" of the paper, the implications are that ~18*10^14 protein search resources would be required for our 18,000 proteins. I'm sure you won't make the mistake of multiplying that for all complex species, as they share many proteins. If our life system currently uses 1 billion proteins, then that would require 10^20 search resources.

forexhr writes:

On the other hand, you need 10^11 evolutionary resources to evolve something so simple as ATP binding protein.

Or any new functional protein in any context. Getting an added step in an already complex system with existing proteins is no more difficult than getting the first proteins in life from random assemblies. Probably easier.

forexhr writes:

But even that simple function is not useful biologically. In the real world, proteins that bind to ATP must also be able to release it to complete their function. That may substantially reduce functional sequences. Szostak and Co didn't test to see if their protein was able to perform that feat in vivo. Their experiment was done in vitro. That is a very large difference. Adding that extra requirement will substantially worsened their 1 in 10^11 ratio.

Really? By how much? Are you suggesting, say, ten times, so it would take 10^21 to get 1 billion proteins? Organisms do their protein searching "experiments" in vivo.

forexhr writes:

But this, of course, isn't something to critically think about. Instead, let us repeat something that nobody disagrees with to prove that those who think critically about the above issues are just stupid creationists.

I'm glad you agree with me so far. So, you can now see that there's no case "against evolution" in your O.P. to be "disproved". There's just a sharpshooter fallacy and technical mistakes.

Still, thanks to you for an O.P. that actually links to some interesting papers, and I think you might thank me for reading the papers and helping you through all this.


This message is a reply to:
 Message 220 by forexhr, posted 04-17-2017 6:34 AM forexhr has responded

Replies to this message:
 Message 247 by forexhr, posted 04-19-2017 4:11 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(1)
Message 246 of 293 (805456)
04-18-2017 4:36 PM
Reply to: Message 221 by forexhr
04-17-2017 6:47 AM


Re: Lambda hits Target with spectacular speed.
Key Innovation in Lambda Phage

forexhr writes:

I don't know what the fuss is all about.

Really?

forexhr writes:

This paper has nothing to do with the issue at hand since it talks neither about evolution of new functional protein folds nor about the ratio of functional amino acid sequences versus the vast combinatorial space.

It didn't take much crossing of that vast space to find a new specific function, did it? You should understand by now from the second paper you linked to in the O.P., the Szostak paper and the one linked to above that potentially functional proteins cannot be very rare islands in a huge sea of non-functionality, as you want them to be.

forexhr writes:

It just talks about the pre-existing viral J protein that acquired the ability to bind a different protein on E. coli, called OmpF, once the LamB protein, to which it normally binds, was turned off. Since both, LamB and OmpF have similar three-dimensional structures, a few mutations in the viral gene fortuitously led to ability of J protein to bind to OmpF. Hence, one out of three functional pre-existing proteins gained the ability to bind a protein similar to one it normally binds, without any change in its native 3-dimensional structure.

It stayed exactly the same while gaining a new function?

forxhr writes:

And this binding ability proves what exactly? That you can produce circulatory system, all joints and bones in your body, or complex, image-forming eyes 100 times independently with 10^43 changes in the spatial positions of molecules?

It is really amazing how you evolution believers explain how everything could have happened: due to few mutations, the pre-existing protein can now stick to something it coud not before - therefore it is a fact that mutations produced all bio-structures that that we observe.

Why is it amazing? If our ancestors observed some rabbits being born from others, why would it be amazing if they inferred that all rabbits came into existence this way? And why would it be amazing if they were sceptical of an alternative suggestion by conjurers that they could produce new rabbits ex nihilo out of hats?


This message is a reply to:
 Message 221 by forexhr, posted 04-17-2017 6:47 AM forexhr has not yet responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


Message 256 of 293 (805575)
04-19-2017 10:15 AM
Reply to: Message 247 by forexhr
04-19-2017 4:11 AM


Re: The Texas sharpshooter rides again
forexhr writes:

I think I already addressed this nonsensical point of yours. Resources that are already spent on search for protein A, which is located in a specific location in the genome of a particular species, cannot be spent on search for protein B which is located in other location of the genome or in other species.

Of course. What I said was this:

bluegenes writes:

Why not just skip to an entire complex organism, like Homo Sapiens. If you "critically examine the logical consequences" of the paper, the implications are that ~18*10^14 protein search resources would be required for our 18,000 proteins. I'm sure you won't make the mistake of multiplying that for all complex species, as they share many proteins. If our life system currently uses 1 billion proteins, then that would require 10^20 search resources.

What I meant by the sentence in yellow is that different species would inherit a lot of the same proteins through common ancestry. Of course they have to search anything they get after divergence separately, and every single unique protein present has to be found at some point.

forexhr writes:

Let me use an example for illustration.

Suppose that each of these two words: "CAT" and "DOG", represents a functional protein fold that needs to be found for adaptation purposes, while the text below represents the genomes of three individual organisms(A, B, C), that all have two duplicated gens, "wdc" and "aii", that are free to explore the above mentioned folds:

A) ..............wdc..................................aii
B) ..............wdc..................................aii
C) ..............wdc..................................aii

After the reproduction, we have these changes(mutations):

A) .....A........wdc...........P........U............Dii
B) .......H......wdc........T.......E................aFi
C) .....T........wdc..........P........Z..............aKi

As we can see, although we have spent 12 mutational resources, we have produced just 3 changes in the "aii" gene, while the "wdc" gene didn't get any change. Since the sequence space of "aii" gene is 17,576(26^3) we need on average 17,576 changes on its locus to produce the "DOG" fold. Once these resources are spent they cannot be spent on search for "CAT" fold on "wdc" loci, or search for any other functional fold. To find the "CAT" fold we need another 17,576 resources.

Well, that was a waste of time. So, if they all have different letters in all positions, it would take 17,576 * 10^9 to get a billion words. You haven't even attempted to refute my actual point.

forexhr writes:

What you did in your nonsensical statement, is put together all the mutations in the history of life that occured in all species of organisms and in all genes, and than you used this number as a separate instance for every new search. This is like writing down every dollar you spent this year and then saying that these, already spent dollars, can now be used for new purchases. All other points in your response are based on misunderstanding of this concept.

No, you couldn't be more wrong. What I did was take the approximate figure it needs to get a target function, 10^11, as the resources to get one functional protein in the life system, and then, by addition, figured out that it would take (10^11)*(10^9)=10^20 to get 1 billion proteins. It's aproximate, but if you understand the implications of the Szostak paper, that's a reasonable conclusion. And it blows out your O.P. claim.

forexhr writes:

Given the fact that you cannot comprehend something so simple(that you cannot spend something that has already been spent), it's no wonder that you bought the theory of evolution from the Darwinian evangelists.

If you could actually grasp the implications of the two papers you linked to in the O.P. and the two I've linked to, you would be thanking me for pointing out where you've gone wrong, rather than continuing to bluster.


This message is a reply to:
 Message 247 by forexhr, posted 04-19-2017 4:11 AM forexhr has responded

Replies to this message:
 Message 257 by jar, posted 04-19-2017 10:27 AM bluegenes has not yet responded
 Message 263 by forexhr, posted 04-19-2017 11:40 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(1)
Message 267 of 293 (805653)
04-19-2017 8:27 PM
Reply to: Message 263 by forexhr
04-19-2017 11:40 AM


Re: please stop B.S-ing and put your arguments on the table
forexhr writes:

Your actual point was even more nonsencical since it used resources needed in order to produce a simple protein composed of 80 AA that performs half-functional binding function, as an explanation for emergence of all complex and structurally independent bio-functions that proteins perform in living systems.

No. I proposed 10^11 for any first protein in the life system, meaning 10^12 for ten proteins and 10^20 for 1 billion proteins. If you disagree with Szostak's estimate, based on searching a random library for an "arbitrary specific function" why don't you write to him and give him the benefit of your expertise?

I also pointed out that the number of proteins required for a highly complex organism with many complex systems, Homo Sapiens, is ~18,000, so it needs ~18*(10^14) protein search resources to get us, step by step over several billion years.

forexhr writes:

bluegenes writes:

No, you couldn't be more wrong. What I did was take the approximate figure it needs to get a target function, 10^11, as the resources to get one functional protein in the life system, and then, by addition, figured out that it would take (10^11)*(10^9)=10^20 to get 1 billion proteins. It's approximate, but if you understand the implications of the Szostak paper, that's a reasonable conclusion. And it blows out your O.P. claim.

You keep repeating the same B.S. about my O.P. claim, and all that you have is the Szostak paper that produced simple binding function with 10^11 resources.

An arbitrary specific function. They could have tested for anything. A functional Lambda tail, for example, something notoriously degenerate.

And that paper is not all I have. Would you like me to give you some more papers so that you can misunderstand them?

forexhr writes:

Can you please finally explain what are those implication? Saying "...If you could actually grasp the implications of the two papers..." is not an argument - this is B.S. ing.

What is it you don't understand about an experimentally supported estimate on how many resources it would take to get any arbitrary specific function?

forexhr writes:

P.S. I predict that your explanation will use already spent resources as a basic premise.

You hope.

Edited by bluegenes, : degenerate spelling


This message is a reply to:
 Message 263 by forexhr, posted 04-19-2017 11:40 AM forexhr has responded

Replies to this message:
 Message 269 by forexhr, posted 04-20-2017 3:14 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(1)
Message 270 of 293 (805674)
04-20-2017 4:03 AM
Reply to: Message 269 by forexhr
04-20-2017 3:14 AM


A Piece of Cake! Progress at last?
forexhr writes:

You proposed a fantasy, something totally unrelated to reality. Life is not composed of 18,000 binding functions without a specific 3D shapes, but of structurally independent and specific 3D shapes. Why would I disagree with Szostak? From what exactly it follows that I disagree? Szostak just showed that you need 10^11 resources in order to produce half-functional ATP binding protein that does not require neither any specific 3D shape, nor the ability to release ATP. You can produce binding function with myriad number of 3D shapes. In a functional sense this is a piece of cake.

If you don't understand what "arbitrary specific function" means and why Szostak uses the phrase, then how can anyone help you?

If you are now agreeing with Szostak that getting function is a 1 in 10^11 "piece of cake", then fine.

forexhr writes:

But when simple binding is not enough and when, in addition to that, you also need some level of 3D specificity to perform bio-function, like in the example of lambda phage genome regulation, then you need 10^63 resources to achieve this function.

No. I told you about superfamilies and how completely unrelated AA sequences can perform the same function, so why are still going on about the 10^63 figure? The authors of the paper in your O.P. certainly don't make that claim.

forexhr writes:

And although both, lambda represor and ATP binding protein, are composed of similar number of AAs(92 vs. 80), due to the 3D specificity requirement, the functional degeneracy is increased by 52 orders of magnitude.

Increased? In which? Where did you get that from? Certainly not from the paper in your O.P.

forexhr writes:

Hence, your fantasy has nothing to do with biology, this is just your personal rationalization to feel comfortable with your dogmatic Darwinism.

Now you're being a bad amateur psychologist as well as a terrible amateur biologist.


This message is a reply to:
 Message 269 by forexhr, posted 04-20-2017 3:14 AM forexhr has responded

Replies to this message:
 Message 271 by forexhr, posted 04-20-2017 4:29 AM bluegenes has responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(1)
Message 282 of 293 (805852)
04-21-2017 7:14 AM
Reply to: Message 271 by forexhr
04-20-2017 4:29 AM


Your capacity for self-delusion is breathtaking
forexhr writes:

bluegenes writes:

No. I told you about superfamilies and how completely unrelated AA sequences can perform the same function, so why are still going on about the 10^63 figure? The authors of the paper in your O.P. certainly don't make that claim.

It is really amazing how you insist on these non sequiturs.

You may not understand the relevance, but that doesn't make my point a non-sequitur.

forexhr writes:

It is true that the relationship between the AA sequence and the 3D structure of a protein is not unique - a large number of modifications in the sequence within a protein family can be tolerated and will result in a similar 3D structure.

And there would also be a large number of functional modifications in any other sequence family that could give you the function. Functionally similar proteins don't need to be identifiable by sequence as family.

Sequence ≠ Structure ≠ Function ≠ Sequence

New Science Press: From Sequence to Function

quote:

As we see next, predicting a protein’s function from its structure alone is complicated by the fact that evolution has produced proteins with almost identical structures but different functions, proteins with quite different structures but the same function, and even multifunctional proteins which have more than one biochemical function and numerous cellular and physiological functions.

forexhr writes:

But that has absolutely nothing to do with resources that are necesary to find a particular 3D structure. If you have 10^57 different AA sequences that result in a similar 3D structure this high degree of conservation of the 3D structure, compared to sequence conservation, does not change the fact that you need 10^63 resources to find this specific 3D structure. Your capacity for non sequiturs is breathtaking.

The 10^57 are only those from one sequential family, not the set of all sequences which would function in that context. And you're still sharpshooting. Like: If one in 100,000 Americans are called Donald Trump, then 100,000 Presidential election resources are needed to get a President called Donald Trump.

Always remember, when considering something like Lambda and its proteins that there's a set of 10^(virtual infinity) of different potential phages with different proteins that have never come into existence.


This message is a reply to:
 Message 271 by forexhr, posted 04-20-2017 4:29 AM forexhr has not yet responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(4)
Message 285 of 293 (805877)
04-21-2017 10:13 AM
Reply to: Message 279 by forexhr
04-21-2017 6:03 AM


forexhr writes:

In my next post I will conclude this discussion with easy-to-understand falsification of evolution.

In my next post, I will give easy to understand explanations of how to produce a rabbit ex nihilo out of a hat*, and how to make a new moon out of Camembert.

*As accurately depicted at left


This message is a reply to:
 Message 279 by forexhr, posted 04-21-2017 6:03 AM forexhr has not yet responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


Message 288 of 293 (806136)
04-23-2017 9:19 AM


Sunday summary: where we are on the O.P. so far.
forexhr writes:

In this thread, I proved that there haven't been enough resources in the history of life to find these 3D shapes.* But, darwinists in this thread completely ignored this important issue, either via non sequiturs, red herrings, appeals to authority, ad hominems and various others pseudoscientific techniques. It is really interesting to watch how pseudoscience operates.

*My yellow. It is indeed interesting to watch how pseudoscience operates.

forexhr in the O.P. writes:

But that begs the question: how did this selectable combination of CHNOPS(protein) came to be? This is the crucial and the most important question. There are virtually infinite number of ways in which CHNOPS comprising protein can be arranged, and most are junk, or non-selectable arrangements. For e.g. for a protein 92 AA long, with 10e122 possible AA combinatios, there is only 1 in every 10e63 functional sequence*. On the other hand, published extreme upper limit estimates puts the maximum number of mutations or CHNOPS re-arrangements at 10e43**. So, the total number of evolutionary CHNOPS re-arrangements is 20 orders of magnitude insufficient to find only one selectable state for evolution to preserve - a protein, let alone molecular machines, organs or organ systems.

*Functionally acceptable substitutions in two alpha-helical regions of lambda repressor. Reidhaar-Olson JF, Sauer RT.

**How much of protein sequence space has been explored by life on Earth?, David T.F Dryden, Andrew R Thomson, John H White

Well, if we wanted to do pseudoscience or bad science, the above shows a good way to start. We could look at an actual 92 AA sequence in one organism, and declare that it would take 10^122 searches of 92 AA sequences (or "CHNOPS re-arrangements") to get it. That, of course, would be wrong, because while every specific random assembly is 1 in 10^122 in search space, the chances of hitting a 1 in 10^122 combination is actually 1 per search, so looking at any functional "hit" that has been incorporated into the life system will give that "1 in 10^122" appearance.

Making one specific observed arrangement into a target with hindsight is an obvious fallacy.

Forexhr doesn't exactly do that creationist classic. Instead of making that mistake, he does it for a specific function. Estimating 10^57 to be the number of possible functional sequences for that specified function, he then concludes that, because they would comprise a 1 in 10^63 part of search space, it would take 10^63 random AA assemblies to get a hit for that function. Once again, any functional 92 AA hit with ~10^57 functional variations will give that impression, so we have the same "target with hindsight" fallacy. The search space would contain many different groupings of ~10^57 that might potentially perform many different functions in a life system, and a life system might hit on some of them, but it certainly doesn't have to search 10^63 times in order to do so. Pick out any one of them after its arrival, and the hindsight probability fallacy can be made, just as with the first example.

To compound that error, forexhr has made a major technical mistake in assuming 10^57 is the approximate limit to the number of AA sequences that could perform his specific function. There isn't actually anything that tells him that in the 1990 paper that he refers to (you can read it from the link above - no paywall as in the O.P.) It shows that there are likely to be at least that many functional permutations of the existing protein for that specific function, but not that there couldn't be many more.

Not only can proteins with very similar sequences perform very different functions, but proteins with very different sequences can perform the same or similar functions.

And proteins with different structures can perform the same function.

There are known examples of this. There are known examples in bacterial phages. There are known examples in the Lambda phage family. There happen to be, hilariously, known examples in the exact protein function forexhr was unfortunate enough to choose for his O.P.

He'd be wise to look it up.


  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(1)
Message 291 of 293 (806329)
04-24-2017 3:11 PM
Reply to: Message 289 by forexhr
04-24-2017 12:16 PM


How many times can someone make the same mistakes?
forexhr writes:

This will be my last post on this thread and in it I will do two things:

Yes. You will make your Hindsight target fallacy, and you will repeat your technical mistakes.

Here's another paper for you to misunderstand, and for anyone who is actually genuinely interested in proteins.

Cro Structure

Instead of running away, why don't you tell us all how American history is impossible because of the impossibility of getting its particular set of Presidents in less than 60 elections?


This message is a reply to:
 Message 289 by forexhr, posted 04-24-2017 12:16 PM forexhr has not yet responded

  
bluegenes
Member (Idle past 675 days)
Posts: 3119
From: U.K.
Joined: 01-24-2007


(3)
Message 292 of 293 (806375)
04-25-2017 8:46 AM
Reply to: Message 289 by forexhr
04-24-2017 12:16 PM


Was this a drive by shooting?
forexhr writes:

To prove my point, in the O.P. I gave the paper whose authors concluded, I quote: "the estimated number of sequences capable of adopting the gamma repressor fold is still an exceedingly small fraction, about 1 in 10^63 of the total number of possible 92-residue sequences."

(Lambda) And that illustrates one of your technical mistakes. "Fold", not "function". As I've pointed out, the authors do not make your claim on 1 in 10^63 for function. That's just your misunderstanding.

Fold ≠ Function.

Cro Structure

forexhr writes:

In other words, there are 10^57 possible functional sequences for lambda represor function and in order to extract one such functional sequence we need on average 10^63 nucleotide rearrangements. Given the 10^43 nucleotide rearrangements in the history of life, it is easy to conclude that the ToE cannot explain the origin of one below average protein.

No. Fold ≠ Function + target fallacy (see below).

forexhr writes:

According to the logic of the average believer in evolution, since nearly all biologists acknowledge that evolution is a fact, the conclusion that the ToE cannot explain the origin of just one simple and below average protein is obviously false. But, given the fact that appeals to the majority are pseudoscientific, the evolutionists on this thread tried to respond to this conclusion through some actual points. These points were all boiling down to this:

1) The sequence space of size 10^122 contains many functional sequences that might perform many different bio-functions (non-Lambda represor functions), and therefore populations don't have to search 10^63 times in order to find them.

The hindsight target fallacy (ex-post facto statistical fallacy)* has been pointed out. A life system doesn't target the existence of Lambda or its repressor. Everything it does get should look very unlikely when viewed as a target, just as the list of American Presidents' names is definitely a less than 1 in 10^122 chance when viewed as a target, and so are the names of your nearest 50 neighbours.

Have you worked out how many functions would have been useful to the Lambda ancestor at the time, and what proportion of 92AA proteins would have provided at least one of them? If 10^112 would have been advantageous, then 10^10 resources should have been ample to add a new functional protein of some kind, and the repressor's ancestor is just as good a result as any other.

forexhr writes:

2) Proteins with very different sequences can also perform the Lambda represor function.

And proteins with different structures/folds. I suspect you might have looked this up, found out that I'm right, and hence the hurried exit. My suspicions could be wrong, of course. You might still be ignorant of the relevant biology.

forexhr writes:

The first point argues that proteins of size ∼92 AA can perform not only Lambda represor function but also many different bio-functions. We can express this point by using an analogy. With three letters we can produce many different meaningful words, and not only words that name animals, for e.g. This is of course true, but it is completely irrelevant to evolution. Why?

Whenever the proponents of the ToE talk about the origin of some bio-function, they will say that this function is simply an evolutionary adaptation by which the organism provides a solution to the problem that the environment sets.

Example:

Meiosis as an Evolutionary Adaptation for DNA Repair
https://www.intechopen.com/...nary-adaptation-for-dna-repair

So, according to the ToE, a particular bio-function is an evolutionary adaptation, while the eonviornment is something to which an organism must adapt.

Given this evolutionary narrative, let us use our three-letter words analogy and set up one population(P) of organisms and its gene pool:

Population P:
organism 1: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 2: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 3: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 4: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 5: -----are -------bit-----hop ----fun------vox----too--- yes---------

So, we have our population and its gene pool which contains 7 evolutionary adaptations(bio-functions).

After the reproduction, an extra copy of a gene is made in the genome of the "organism 2", which gives us the following gene pool:

Population P:
organism 1: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 2: -----are -------bit-----hop ----fun------vox----too--- yes---wox-
organism 3: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 4: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 5: -----are -------bit-----hop ----fun------vox----too--- yes---------

This duplicated gene is free to explore new adaptations(bio-functions). But, given the evolutionary narrative, in order for an organism to adapt, we also need an environment. So let us suppose that the following new environment emerges:

"All three-letter words that name animals".

In the contect of this environment this adaptations are functional:

ant, ape, auk, bat, bee, bot, boy, bug, cat, cod, dab, doe, dog, eel, eft, elk, emu, ewe, fly, fox, gnu, guy, hen, hog, jay, kea, kit, owl, pig, ram, rat, sow, teg, cow

Since the sequence space of three-letter words is 17,576(26 letters ^3) and we have 35 functional sequences, it follows that only 1 in 502 three-letter sequences is functional. In other words, on average we need 502 resources in order to adapt to our new enviornment.

Now, given the first point made by evolutionists on this thread, they would respond in the following way:

"Forexhr, you made a major technical mistake in assuming that you need 502 resources to adapt to "All three-letter words that name animals" enviornment, because the Scrabble Dictionary recognizes 1015 three-letter words that are also functional."

The above claim presupposes that because theoretically we can create 1015 different "functions" with three-letter words, it follows that we don't need to spend 502 resources in order to find "functional" three-letter words that name animals, but only 17 (17,576/1015).

No. If the animal words represent the only words which are useful to the organism at the time in your analogy, and a new coding gene arrives, 502 resources would be right. If other 3 letter words could be useful at the time, then it's less than 502 for any new gene. Whatever does arrive is not a target. So, how many different non-repressor functions would have been useful to the Lambda ancestor when it first gained its repressor? How many different evolutionary paths could it have taken without a repressor but with something else? How big is the field of potentially functional descendents of that Lambda ancestor that never came into existence? 10^1000 or more?

forexhr writes:

The absurdity of such reasoning is obvious. We cannot use all actual, or potential bio-functions that were, or will be produced with 92 AAs and then claim that the existance of these bio-functions reduces the resources necessary to adapt to a particular environment.

But how have you worked out what proportion of 92 AA proteins would have been of some use to Lambda's ancestor in any function at the time, and how do you know what actual function the repressor's ancestral protein had, and what makes you think it was necessary on arrival rather than just one of many possible advantages?

forxhr writes:

Given our population P:

organism 1: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 2: -----are -------bit-----hop ----fun------vox----too--- yes---wox-
organism 3: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 4: -----are -------bit-----hop ----fun------vox----too--- yes---------
organism 5: -----are -------bit-----hop ----fun------vox----too--- yes---------

The fact that the three-letter words - "are", "bit", "hop", "fun", "vox", "too" and "yes" exist in the gene pool of the population, cannot change neither the sequence space of three-letter words(17,576), nor the number of three-letter words that are required for adaptation to the current environment("All three-letter words that name animals") at the "wox" loci.

So, the first point is pure nonsense.

So, if 2*(10^119) 92AA proteins would have been useful in some way to the Lambda ancestor at the time it received its repressor, that would match your analogy, and it would take ~500 resources with the repressor being just as good a result as any other, like "rat" in your analogy.

forexhr writes:

The second point, which argues that proteins with very different sequences can also perform the lambda represor function, is like saying that we can use very different three-letter words to fulfil the above mentioned ecological niche - "All three-letter words that name animals".

As such, this point is just redundant rhetoric because if 10^57 possible sequences for Lambda represor are functional that also means that many of them are "very different". But of course, 1 in 10^63 ratio still stands.

Fold ≠ Function and a specific function isn't a target.

forexhr writes:

Likewise, in the group of three-letter words that name animals, many of them are very different(fox vs. gnu for e.g.), but from that it doesn't follow that the above mentioned ratio(1 in 502) is not valid.

But you've made them both up. Your 1990 paper doesn't claim 1 in 10^63 for function and no function is a target. Because you don't know what proportion of all structures might perform a specific function, the best thing to do is to test random strings for an arbitrary specific function (Szostak: ~10^11 for a hit).

Functional Proteins from a random sequence library

forexhr writes:

As we can see, the responses given are totally unfounded, they are just nonsensical ad hoc excuses whose only purpose was to keep faith in the ToE and to have some justification for labeling my argument a fallacy(sharpshooter for e.g.). Therefore, my statement, that evolution can't even explain the origin of one below average protein is correct.

Since convincing an evolutionist that the ToE is false is like convincing a flat-earther that the Earth is spherical, there's nothing here for me anymore. Hence, over and out.

I think you're running scared, and might have found out that Fold isn't Function.

*I thought I'd just add a bit more pretentious Latin to the thread.


This message is a reply to:
 Message 289 by forexhr, posted 04-24-2017 12:16 PM forexhr has not yet responded

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2018 by EvC Forum, All Rights Reserved

™ Version 4.0 Beta
Innovative software from Qwixotic © 2019