Biophysicist Kirk Durston of the New Scholars Society offers an explanation below:
Note: Kirk Durston is back and this post has been stuck to the front page until late this evening EST, so that further comments and his responses may be noted. – News
–
There is a common either-or mistake made by most Darwinists and, quite frequently, by ID theorists as well. The mistake, which is an example of the fallacy known as the false dichotomy, can be described as occurring when one assumes that either no functional information encoded in the genomes of life can be produced by natural processes, or all of it was produced by natural processes. A closely related mistake made by Darwinists is the assumption that if natural processes can produce a trivial level of functional information, then we can safely conclude that natural processes can produce all biological information.
There are still challenges in mathematically defining functional information or functional complexity. For my purpose here, however, I will simply use the mathematical descriptions published by Hazen et al., and Durston et al. Both approaches cited are very closely related to an earlier equation published in 1951 by Leon Brillouin, which can be simply represented as
FI = -log nf/nt
Where nf = number of sequences that are functional and nt = the total number of possible sequences. It should be clear from the above equation that if nf is large enough for examples to be generated by random recombinations, then functional information (FI) can be generated by random natural processes, albeit a trivial level. For example, it is clear from work done at the Georgia Institute of Technology, that nf for simple binding pockets is pretty high, which entails that the FI required to code for binding pockets is relatively trivial.
Reflection on the above equation reveals that the FI required for a given function can range anywhere from zero to some very high number. It is, therefore, a mistake to assume that FI can only be generated by intelligence; a trivial level of FI can be produced by completely mindless processes, as should be obvious from the above equation, and as the Georgia Tech results illustrate.
It is also a mistake to assume, as many Darwinists do, that because mindless processes can generate a trivial level of FI, therefore mindless processes can generate high levels of FI. Again, reflection upon the above equation (or the more detailed equations published by Hazen or Durston) reveal that the higher the FI required, the less probable it becomes (i.e., the nf/nt ratio approaches zero).
The fatal mistake made by Darwinists at this point is to invoke what has become the Darwinist god-of-the-gaps, namely selection. As we can illustrate from evolutionary algorithms, selection requires a fitness function which, itself requires FI to encode. Of course, it follows from what I am arguing here that trivial levels of selection can be produced with trivial levels of FI. The question is whether natural selection has sufficient information to locate stable, functional, biological proteins. All our work to date seems to falsify that option and verify the need and actual role for intelligent design (in this case human) when producing artificial proteins of any significant structure. To clarify, recent building of artificial proteins is an example of intelligent design in action.
The Georgia Tech work has led some Darwinists to believe that because binding pockets are relatively trivial to encode in a sequence that, therefore, we have somehow explained how natural processes could have encoded biological proteins. In real life, however, proteins are about a lot more than simple binding. Binding to the right molecule is important, at the right time, at the right location and with the right binding strength so that the bond can be broken at the right time and place, etc. This can often require a larger 3D structure for proper functionality, that has a nf/nt ratio approaching zero. For example, if we take the results published for 35 protein families by Durston et al., and solve for nf/nt, we observe that it is extremely small for many protein families.
My contention is that the ability to generate statistically significant levels of functional information is unique to intelligence. It follows from this that if a function can be achieved with a statistically insignificant level of FI, then intelligence is not required. Statistical significance, therefore, is the safeguard against false positives and can be measured in a variety of ways, such as measuring the adjusted residual of the outcome and choosing a cutoff that represents a very high confidence level, such as 99.9% or greater. With this in mind, an executive summary of my own case for intelligent design in biological life is available here.
Related note as to the Google Tech work and how it relates to reality:
Why Skolnick and Gao (2013) Doesn’t Refute ID Research on the Origin of Proteins
Casey Luskin July 23, 2013
Excerpt: (1) Skolnick and Gao don’t cite any work of ID proponents nor do they claim their paper pertains to it.
(2) Skolnick and Gao’s work is entirely theoretical and computational, and is thus hard-pressed to impinge upon Axe’s experimental results.
(3) Many other researchers working on protein structure disagree with their conclusion, and believe “there are many more than 500 unique sites among the human proteome.”
(4) Skolnick and Gao only studied the behavior of an artificial set of proteins which were assumed to already be capable of folding into a stable structure, and thus their paper doesn’t address Axe’s research which investigated the likelihood of producing stably folded proteins in the first place.
(5) If Skolnick and Gao are correct that proteins are clustered closely in sequence space, it still doesn’t show that closely related proteins can always evolve from one function to another: Axe and Gauger (2011) experimentally found that even with proteins very close in sequence and structure, moving from one function to another can require more simultaneous mutations than could arise through a Darwinian evolutionary process over the entire history of the earth.
(6) Skolnick and Gao only studied proteins that bind small molecules like ligands. They did not study the feasibility of evolving much more complex protein shapes required for many common enzyme functions, like enzyme catalysis, involving much larger and more complex molecules.
(7) Their paper attempts to theoretically bolster the “promiscuity hypothesis” of protein evolution, a hypothesis that has not fared well in experimental tests.
Skolnick and Gao ask, “How can one demonstrate that the ability to engage in a variety of low-level biochemical functions without selection is an intrinsic property of proteins… ?” In other words, what should convince us that the functions of natural proteins are abundantly present in random polypeptide chains? The answer is obvious: Go into the lab and impress us with what you find in a mixture of random polypeptides.
Neither they, nor other critics, have done this.
http://www.evolutionnews.org/2.....74761.html
correction: Related note as to the Georgia Tech work and how it relates to reality:
sorry guys, I really need to get into the habit of giving my posts the once over before I submit.
This is pretty damn typical of IDists. They always ask for evidence that natural processes can do the trick and when they are shown it, they stick their fingers in their ears. “It’s just a flesh wound”, they cry. Here is an experiment that shows biological information can be produced by nature, which they have long asked for, an it;s not good enough. The line gets pulled back and they want evidence that nature can produce REAL information, not just any information. We just haven’t found how yet but it’s coming. It stands to reason that if nature can produce information, amino acids, nucleotides etc then somehow it produced the first cell with all of this. We will figure it out soon. I would suggest that the IDists here start sticking your fingers in your ears even deeper now to save you the trouble in the future.
BA77, you always ask to show how nature can produce proteins from scratch using blind processess. Is this good enought for you or are you going to move the goalposts once more? Face it, my freind, you and your theology are losing. Give it up.
I must thank ID though. If it wasn’t for ID, the good science might not have worked as hard to refute the bad science.
Not at all. To produce rocks and to generate cathedrals made of rocks are two different things. That nature produce the first cell from molecules is yet more difficult than that.
Durston says that natural processes produce *trivial* FI and that real proteins contain far higher levels of FI. Moreover from proteins to organisms there is a deeper gap than from rocks to cathedrals.
The bottom line is that Darwinian molecules-to-man is pure absurdity.
JLAfan2001-
There isn’t any evidence that blind and undirected chemical processes can construct multi-protein configurations. And seeing tat living organisms contain many of those it appears that your position can’t explain jack.
Look at Lenski’s long-running experiment- no new proteins were produced.
You have a better chance of showing that nature produced Stonehenge than you do of showing nature can produce a living organism from non-living matter.
JlAfan2001 you claim,
And yet Casey Luskin, in the article I I listed directly above your post (which apparently you did not read), points out that ‘nature’ had nothing to do with the ‘experiment’:
Ummm JLAfan2001, not to doubt your unflinching faith in all things Darwinian, but if intelligently designed computational models/programs are required to explain the origination of even ‘trivial levels of functional information’ (since no one has actually seen it happen in real life) then shouldn’t this at least raise a red flag in your book that all may not be well in Darwinland? Especially since the level of functional complexity that needs to be explained in the real world, in real living organisms, greatly exceeds the functional complexity of any computer program man has ever devised, much less these ad hoc computer programs that were intelligently designed to demonstrate the origination of ‘trivial levels of functional information since Darwinism cannot offer a real world example??
Notes as to the sheer disconnect between the empirical evidence (of the real world) and what needs to be explained:
JlAfan2001, perhaps you have no problem using intelligently designed computer programs to demonstrate that unguided Darwinian processes can produce computer programs of unfathomed complexity, but others not so enamored with all things Darwinian, might question your sanity for claiming as such! Just saying!
JLAfan
The issue isn’t just producing proteins, but how undirected, natural causes produce proteins the exhibit complex, specified information. Perhaps you could reference a single peer reviewed scientific research study that demonstrates how undirected, natural causes accomplishes this feat. (hint: there isn’t one…not one!)
If this was something to be fluffed off as trivial, why is Kirk Durston and Fazale Rana concerned with it? They both seem to think that this isn’t as trivial as ID thinks it is. They acknowledge that some arguments and a different approach should be taken in defense of the idealogy. In other words, hand wave away the evidence and change the arguments rather than just accept that they’re wrong. DonaldM, the OP shows the evidence that it can be done. It may be using a computer program but so what? Doesn’t ID use computers to simulate any experiments or has it all been natural? This just proves my point. It’s never good enough evidence for ID. It HAS to be a certain way, in a certain time, in a certain lab, with a certain person, doing a ceratin thing, producing a certain result, using a certain method otherwise it won’t be believed. ID wants the impossible as evidence before it gives up it’s faith.
JLAfan2001 is upset because his position doesn’t have any evidentiary support- heck it can’t even muster testable hypotheses.
i study biology and evolution
:i have a very strong evidence for design in nature
a) we know that a self replicate robot that made from dna need a designer
b) the cat\ape is a self replicate robot
a+b= the cat need a designer
?plus: if a self replicate car cant evolve into an airplan, how can a bacteria can evolve into human. and if we dont have a step wise from a self replicat material’ then we cant get it in nature(a minimal motion system need minimum 2-3 proteins
about the similarity in 2 animal argument: a 2 cars of honda can look very similar to each other. but this is because they made by the same designer- honda company
according to the evolution a poo(colection of bacteria )can evolve into a princes
Joe
You’re the poster boy for my statement above. Always wanting something extremely specific in order for him to change his mind. Then claims that the people he opposes are close minded. Show me on natural thing by experimant in the lab or in the field that a designer was invloved in. I don’t want human made examples and I don’t want evidence after the fact. Show one experiment that we can see a designer doing something in real time. If you can’t then your position is just as oblivious as the position you oppose.
JLAfan2001 you state:
What point? That if you ignore the computer programmer behind the curtain you can live in ‘The Wizard of Darwinland’?
Notes as to refuting (intelligently designed) evolutionary computer programs:
Notes as to using realistic parameters trying to simulate evolution in computers:
Or as Eric has put it:
The question is what is “natural” i.e. how did nf come to be large enough to allow trivial random processes to produce FI. ID variants that accept such framing with undefined, open ended “natural”, which amounts to signing a blank check, cannot but keep backing from the ever expanding domain of “natural” since at the end everything that is will be shown to be “natural” (universe being optimized to be knowable).
The only tenable ID position is that the whole nature is the result of design and continued computations of its unfolding, from ground up and at all points in space-time, not from some shifty boundary and up as the present part-time-no-tomorrow ID gratuitously concedes.
ID computer programs = good and unbiased. Darwinist computer programs = bad and biased. You can trust ID to be a simlulation of reality and honesty but you just can’t trust the evil, conspiracy motivated skynet computers of the Darwnists. They are programmed to deceive and disprove a designer.
JLAfan2001, I’m with you 100%, let’s throw all these bogus computer simulations out the window and see what the real world can tell us about Darwinian processes, since empirical evidence is suppose to have final say in science anyway?
OOPS!
JLAfan2001,
You said in #3:
From the OP:
Thanks for the nice illustration of exactly the mistake that Dr. Durston pointed out.
Yes JLAfan2001, I understand that you are upset because your position has absolutely nothing.
Too bad for you…
JLAfan2001, I’m leaving in an hour for several days vacation, so it may be a while before I can continue my response here. First, there is no ‘pulling back’ the line. I have argued for more than 20 years that there is a threshold below which any functional information is trivially easy to produce. You might be able to dump a bunch of letters on the floor and accidentally generate the word ‘be’ but it would be irrational to think one can write a meaningful essay that way. In the same way, it is irrational to think that because natural processes can generate very low-level, trivial functional information within the background noise of natural processes that, therefore, we have explained where the information came from to generate a stable, folded and functional protein family. Saying that it ‘stands to reason’ that if nature can produce a trivial level of information then it can produce a cell lacks credibility. If you want to post a credible response, you will need to actually show that ‘it stands to reason’. The way you can do this is to read the papers I referenced, understand the equations, then start crunching the numbers and do the work. I’ve done the work for several protein families and published the results in my paper cited in my post above. You need to do the work. Simply asserting that nature did it is bad science. In fact, it is not science at all. Perhaps an easier project for you or others would be to attempt to falsify the hypothesis I present in the executive summary linked to at the end of my post. Science is about hypotheses, predictions and falsification. My hypothesis makes predictions that can be falsified, but you will need to do the work. Simply making an assertion or expressing a belief does not falsify a hypothesis. No amount of assertions is going to do the job; you need to do the work. Until you or someone else falsifies my hypothesis, my hypothesis stands, along with its conclusions.
I an extremely happy of this thread. Durston has given us, again, a very clear exposition of the true foundation of scientific ID theory. As usual, he is completely correct and his concepts are simple and true.
The method devised by Durston to indirectly measure FI in protein families remains the best operative approach to understand and analyze FI in a biological context.
Any intelligent and unbiased person should recognize the importance of this work.
Here is the reference information for some the articles mentioned in my initial post:
Hazen RM, Griffin PL, Carothers JM, Szostak JW (2007) Functional information
and the emergence of biocomplexity. Proc Natl Acad Sci U S A 104 Suppl
1:8574-8581
Durston KK, Chiu DK, Abel DL, Trevors JT (2007) Measuring the functional
sequence complexity of proteins. Theor Biol Med Model 4:47
Brillouin L (1951) Physical Entropy and Information. II. Journal of Applied Physics 22:338-343
Jeffrey Skolnick and Mu Gao, “Interplay of Physics and Evolution in the Likely Origin of Protein Biochemical Function,” Proceedings of the National Academy of Sciences, USA 110 (June 4, 2013): 9344-49.
Dr. Durston, as well, I like your latest paper on proteins since it kicks the whole debate up a notch to ‘context dependency’ of AA sequences:
Statistical discovery of site inter-dependencies in sub-molecular hierarchical protein structuring – Kirk K Durston, David KY Chiu, Andrew KC Wong and Gary CL Li – 2012
Results
The k-modes site clustering algorithm we developed maximizes the intra-group interdependencies based on a normalized mutual information measure. The clusters formed correspond to sub-structural components or binding and interface locations. Applying this data-directed method to the ubiquitin and transthyretin protein family multiple sequence alignments as a test bed, we located numerous interesting associations of interdependent sites. These clusters were then arranged into cluster tree diagrams which revealed four structural sub-domains within the single domain structure of ubiquitin and a single large sub-domain within transthyretin associated with the interface among transthyretin monomers. In addition, several clusters of mutually interdependent sites were discovered for each protein family, each of which appear to play an important role in the molecular structure and/or function.
Conclusions
Our results demonstrate that the method we present here using a k-modes site clustering algorithm based on interdependency evaluation among sites obtained from a sequence alignment of homologous proteins can provide significant insights into the complex, hierarchical inter-residue structural relationships within the 3D structure of a protein family.
http://bsb.eurasipjournals.com/content/2012/1/8
(A Reply To PZ Myers) Estimating the Probability of Functional Biological Proteins? Kirk Durston , Ph.D. Biophysics – 2012
Excerpt (Page 4): The Probabilities Get Worse
This measure of functional information (for the RecA protein) is good as a first pass estimate, but the situation is actually far worse for an evolutionary search. In the method described above and as noted in our paper, each site in an amino acid protein sequence is assumed to be independent of all other sites in the sequence. In reality, we know that this is not the case. There are numerous sites in the sequence that are mutually interdependent with other sites somewhere else in the sequence. A more recent paper shows how these interdependencies can be located within multiple sequence alignments.[6] These interdependencies greatly reduce the number of possible functional protein sequences by many orders of magnitude which, in turn, reduce the probabilities by many orders of magnitude as well. In other words, the numbers we obtained for RecA above are exceedingly generous; the actual situation is far worse for an evolutionary search.
http://powertochange.com/wp-co.....Myers_.pdf
KD
I’m willing to see your point and I concede that I don’t know enough of this area to disprove it. I will have to leave this to others who do enough and can do the relevant heavy lifting.
This seems a very odd comment
Encode where?
Is this just a fancy way of saying that Darwinists can’t account for OoL?
Because once you have Darwinian-capable self-replicators, then that’s the coding done.
Sure, that’s a big ask, but it’s not a Darwinian ask, it’s an OoL ask.
Elizabeth’s bald assertion:
So you say yet cannot demonstrate. IOW all you have are your bald assertions and it is strange that you think they are actual evidence.
No, it is a straightforward way of saying that (even if handed self-replication) Darwinists usually can’t account for the information they smuggle into their fitness functions.
Elisabeth stated on her blog that she is not a ‘Darwinist’ *if* that term involves ‘ideology.’ It is quite obvious, however, that most people here at UD think it involves ideology and not just science. Yet it is strange that Dembski calls ‘Darwinism’ a ‘scientific theory’ (2004).
This is perhaps why Durston commits ‘scientism’ without even knowing it with his ‘strictly scientific’ rendition of Big-ID Theory = science-alone. (That’s why he doesn’t capitalise ‘intelligent design’ qua theory in his writing; he doesn’t want you to know that committment to scientism is what he is actually doing to IDism.)
It may also explain the Darwin-phobia he demonstrates in this thread, starting the 1st, 5th, 6th and 7th paragraphs specifically focussed on his bogey-men and women. It might surprise him how few people nowadays seem willing to accept the label ‘Darwinist’ as science moves onwards, even if sometimes it is difficult (especially in N. America) to tease apart the ideology from the science.
Just curious: Is ‘www.powertochange.com’ an evangelical, Christian apologetics site, where Durston posted his bait-and-switch so-called “Scientific Case for Intelligent Design”? The “New Scholars Society” certainly is.
It may seem totally irrelvant and even unfair to some of you (especially if you are of the same fold) to point that out. The complaints against such facts are too ironic, however, to miss. It is as if being an actual person, according to this school, has *absolutely nothing* to do with whether or not one accepts Big-ID qua scientistic ideology simply because they/you have a mantra to offer – ‘following the evidence where it leads.’ Yet in this case, follwing the evidence leads directly from IDism to evangelical Protestant Christianity via Durston’s link, so that evidence should ‘speak for itself’ too.
And yes, David Berlinski is still an ‘agnostic’ (according to News) and therefore cannot offer a positive case for Intelligent Design like Durston does on the basis of his personal evangelical Protestant theology (oops, it should read: on the basis of his impersonal, dispassionate, neutral science). Sadly, just like the atheists they seek to oppose, in cases like Durston’s they are actually contributing to the science vs. theology/worldview ‘warfare’ model, rather than helping to overcome it.
Kirk, in case you are too ready to willfully misinterpret me, to call me a Darwinist or something I’m not; I’ve been pointing out calmly and clearly for several years that ‘Intelligent Design Theory’ should be properly understood as a topic for science, philosophy and theology dialogue. Iow, integrally instead of fragmentarily as IDism still does.
Gregory:
blah, blah, blah…guilt by association…blah, blah, blah
I’d be more worried about Kirk’s readiness to willfully ignore you, notwithstanding narcissistic assumptions to the contrary.
Are genetic algorithms an example of a mindless process?
Elizabeth’s other bald assertion:
sez who? Elizabeth Liddle? And how does she know?
Tell us Dr. Liddle, what is the minimal system required for Darwinian evolution? Is such a system “alive,” and if so, according to which definition of “life”?
Are genetic algorithms alive? Are they capable of Darwinian evolution?
I have a question for KD when he gets back..or anyone else for that matter.
Consider a complex protein with a great deal of FI. Lets ignore for the moment how this protein came into existence. Durston suggests that small amounts of FI can be produced by evolution. Could natural selection add the FI of the complex protein by some incrementally small amount?
If so, why. If not, why not?
Of related note:
A Response to Martin Poenie on Protein Evolution – Jonathan M. – July 25, 2013
http://www.evolutionnews.org/2.....74821.html
Elizabeth:
While I cannot certainly speak for Durston, I would like just the same to comment about your post #23.
a) First of all, it seems rather odd that you try to separate so strongly two problems that are, indeed, strictly connected. OOL needs living beings to come into existence, and living beings do need a lot of biological information to exist. New species have similar requirements: a lot of new biological information. Even if we keep it proteins (which certainly are not the whole story), new original proteins are needed both for OOL (certainly at least a few hundreds of them) and for the emergence of new phyla, orders, species, and so on. The two problems are obviously similar, and it would be very strange indeed that they require two completely different solutions.
b) The neo darwinian algorithm, as clearly stated by Durston, and in general by ID theory, can certainly generate new functional information, but that is always simple information, that can arise by random variation, and be simply selected if the conditions allow for that. But the capacity of the neo darwinian algorithm to generate some new information are strictly a consequence of the huge original information that already is in living replicators. IOWs, very complex functional objects, like living beings, can occasionally benefit of some random “tweaking” of the existing information, but that is never the emergence of new complex solutions. That’s what we can observe (microevolution). Macroevolution (the emergence of new, original, complex functions that require hundreds of bits of coordinated functional information, is definitely beyond any reasonable causal power of the neo darwinian algorithm. That’s why no instance of neo darwinian macroevolution has ever been observed, or even hypothesized with some detail, at the molecular level.
So, to sum it up: the problem is the same, both for OOL and for evolution of the species. In both cases, the neo darwinian algorithm (or any other algorithm which does not include intelligent intervention) cannot explain the huge quantity of complex functional information that we observe routinely in biological beings. Indeed, not even the emergence of a new functional protein of average length.
I made a small change in your quote so it made sense to me.
It is better to say allele rather than protein because one insertion or deletion can change the complete sequence of amino acids in the protein. I am certainly not an expert on proteins but I don’t think anyone who espouses ID think that this is not possible or does not happen.
As an example, a change in one nucleotide could cause a color change in the fur of an animal. The gene pool is now larger and there is more information in the gene pool. I believe Behe, in the Edge of Evolution, described a frame shift in an allele for certain type of fish produced an anti-freeze protein that enables the fish to survive in sub-freezing temperatures.
So I am sure that other small changes could lead to some functionality. Trivial changes are not denied. But one cannot run with a trivial change and assume more complex changes. Complex changes are not the accumulation of trivial changes.
The question is can there be a series of changes to some part of the DNA that would end up with an allele sequence that codes for a protein unlike any other protein in the organism and it is useful. I believe Durston is saying no.
Do these DNA changes happen? Apparently so. There are known processes that create non coding sections of DNA that are free to mutate in a lot of different ways but the probability of one leading to a protein with functional folds is extremely low, the probability of this process leading to two that coordinate with each other is infinitesimally low and many systems in an organism require not just a couple but often several proteins to function together in a very intricate and coordinated way. So from what I understand it is a probability issue and not just a building on one trivial change after another. Yes, anything is possible but what are the odds that it will happen?
Not very likely even if the machinery to do this creation is set up already. And this is even a bigger probability problem. The translation and transcription machinery is much more complex than a typical protein system yet it is just blithely assumed to happen so the other incredibly process can happen.
I always maintained that this machinery to allow organisms to adapt in these small ways for survival in incredibly amazing design. In other words what Darwin saw was design working to allow organisms to adapt to different environments. He did not see the accumulation of small changes leading to completely different organisms. That was a leap of faith that had no basis in any science we know of then or today.
RodW (#30):
I am not sure I understand well your question, so I will try to answer to two different possible meanings of it:
a) If you are asking if, once a complex functional protein exists, RV and NS can modify it adding some new functional information, the answer is: yes, certainly, provided that the functional variation generated is simple enough.
We can certainly discuss what is the real threshold of functional variation that is in the range of RV + NS. I believe that Behe puts it at about 2 or 3 coordinated AA mutations, starting from merely empirical observations. Axe, if I am not wrong, gives a maximum threshold of 5 or 6 coordinated AA mutations, reasoning from a more general system of observations and considerations. I strongly believe that they are very near to the truth, and that the real maximum threshold of functional variation that RV + NS can (exceptionally) achieve is somewhere between 3 and 5 coordinated AA mutations.
And I think I am very generous here. Indeed, all that we can observe tells us that single aminoacid functional mutations are an event that, although rare, is common enough in the long run. Two aminoacid funtional mutation may probably occur, but they are truly exceptional. Three two five, we are in the range of possible, but so rare that it cannot really have a substantial role in our problem.
As anyone can see, we are however extremely distant from an even vague possibility ton explain the emergence of a new functional protein.
b) If, instead, you are asking if RV + NS can be responsible for the gradual emergence of the protein itself by some incrementally small amount, then the answer is definitely: no.
Why? Because, as the number of coordinated AA mutation that are necessary to generate the new functional information increases, the search space increases exponentially, and the probabilities of getting some instance of the target space (the functional subset of sequences) by RV fall very early well beyond the probabilistic resources of any known biological system, including our planet with its whole natural history.
As I have said many times, that model of “incrementally small amounts” could work only in one scenario, which is obviously false: if each small amount of new information were naturally selectable. And that is simply not true.
Let’s make an example. At some point in natural history, a new functional protein, let’s call it “B”, emerges in a new species. The new protein is the first instance of a new protein superfamily. IOWs, it has no sequence homology with anything that already exists, it has a new fold, and it has a new biochemical function. Please, note that such an event has occurred hundreds and hundreds of times during natural history, after OOL.
Now, let’s say that protein B is 150 AAs long (a small protein, certainly). And let’s say that it emerges by mutations from some unrelated part in the genome (It does not matter if it was a duplicated unrelated gene, or some trait of non coding DNA, or anything else).
As the search space for a 150 AAs protein is 20^150, we can easily understand that, even if the target space if big, we can never expect that a random walk will get us to the target space. If you are interested, the Durston method allows us to quantify with more precision the functional information necessary to express some definite biochemical function (see his evaluation of the FI in 35 different protein families).
Noe, to achieve our observed result through RV + NS, we have to assume that each simple mutation, for example of one AA, generates a functional state that can be naturally selected. IOWs, each AA mutation should give a sequence that is more functional than the previous state to the point of conferring a reproductive advantage, so that it can be positively selected. Moreover, each of those selectable states must be an intermediate to the final functional protein: IOWs, it must be nearer to the final sequence than the previous state.
To make it more clear, each time that one AA mutates, two different, and unrelated, things must be true:
1) The new sequence, with that single aminoacid variation, must confer a reproductive advantage (whatever the reason for that).
2) The mutated aminoacid must the right one at the right place for the future functional protein (whose function has not yet emerged).
If anyone can believe that all that is possible, and that it happened hundreds of unrelated times, then he is on the right way to be a good darwinist 🙂
JLAfan2001:
It stands to reason that if nature can produce information, amino acids, nucleotides etc then somehow it produced the first cell with all of this.
It stands to reason? Why?
Please, read my previous post to RodW. I will be happy to discuss your comments.
I’m willing to see your point and I concede that I don’t know enough of this area to disprove it. I will have to leave this to others who do enough and can do the relevant heavy lifting.
Ah, I understand. Then, why don’t you also leave to those other “heavy lifters” the responsibility to make statements here? It’s quite easy to make statements, and than say that you have not the knowledge to support them…
RodW-
It is all about EVIDENCE- as in there isn’t any evidence taht natural selection can do the job required.
Just look at Lenski’s long-running experiment. Nothing there that helps you.
How living organisms came to exist is directly related to how they evolved. If the OoL was designed then it follows that organisms were designed to evolve and evolved by design.
It is only if the OoL occurred via blind and undirected processes can we infer darwinian evolution is responsible for its subsequent diversity.
That evos refuse to understand that simple fact exposes their agenda.
gpuccio
But we do not know this – it’s the subject matter of OoL research. We do not know how simple a self-replicator has to be to be Darwinian-capable. It might be very simple indeed, and within reach of chemistry and physics, or it might not be.
But we can’t assume one way or the other a priori.
In addition, we can’t assume that the simplest Darwinian-capable selfreplicators had proteins at all. So there is no a priori reason for saying that Darwinian processes can’t have led to functional proteins, and that therefore the only non-design pathway was random draw.
I think this is a misunderstanding. There’s nothing special about phyla – all lineage divergences, under the Darwinian hypotheses, are speciation events, and speciation can happen without radically new proteins emerging. And we simply do not know that “new original proteins” are needed for OoL.
It’s possible that peptides were part of the pre-biotic scene, but possibly not. It’s a question worth asking but not one to which we can assume we have an answer.
They must require completely different solutions, because you can’t invoke Darwinian processes to account for the first Darwinian-capable self-replicator! Whereas you can invoke Darwinian processes to account for things that evolved as a result of there being Darwinian-capable self-replicators.
So the interesting OoL question is: how simple did that first Darwinian-capable self-replicator have to be, and (for IDists), was it within reach of non-Darwinian processes, such as the physical and chemical processes on early earth?
Elizabeth:
Lizzie, scioentists have been working on this issue for many decades. Therefor if it is simple then you are calling all of those scientists imbeciles.
You are still living in a fantasy world and confusing it with the real one.
Evidence, Lizzie, you don’t have any evidence to support anything you say.
Nope, blind and undirected chemical processes are still blind and undirected chemical processes.
Also Tracy and Lincoln’s experiment with a self-sustained replication of RNAs refute your claims. I take it that is why you ignore their results and prattle on anyway.
My apologies for not proofing my post. As it is my question is vague..so I’m reposting here with correction:
I have a question for KD when he gets back..or anyone else for that matter.
Consider a complex protein with a great deal of FI. Lets ignore for the moment how this protein came into existence. Durston suggests that small amounts of FI can be produced by evolution. Could natural selection add to the FI of the complex protein by some incrementally small amount?
If so, why. If not, why not?
Elizabeth,
How Dr Nim came into existence directly affects how it operates from that point on.
Go figure. The same can be said for computers, automobiles, TVs, blenders, microwaves, etc.
Elizabeth:
We have been there before…
Yes, but “a posteriori”, from all that we know, the only self-replicators that have ever been observed in the natural world are, at present, prokaryotes. You are free to imagime “a priori” any possible darwinian self-replicator, be it stone, air or RNA, but as far as I know there is nothing that can help those myths to “evolve” from their status of pure imagination to the status of scientific hypothesis.
Please, refer to the previous point. I am trying to discuss science here. Although I like fantasy fiction, this is probably not the place to discuss it. By the way, I have just begun the “Wheel of time” series! 🙂
It’s not a misunderstanding. I am not interested in phyla or species here. New proteins simply emerge. That is a fact. They emerge in new species (or phyla, or whatever), and they do have new sequences and new functions.
At OOL (let’s say in LUCA, that for me, and for all reasonable people, is the only OOL empirically known), less than half of protein superfamilies were present. The others emerged after that. Are you denying this simple fact?
An answer that we certainly do have is that functional proteins, hundreds of them, were part of LUCA, the only OOL empirically known. For me, that is a very interesting answer, and not one that is based on pure imagination.
Must? That is funny indeed! I will never invoke darwinian processes to account neither for the first living replicators nor for for what evolved after. I still have some respect for reason and science. I maintain That OOL and the following evolution of species should suggest the need for a similar explanation to all reasonable people who are not committed by faith to give unwarranted room to the neo darwinian theory “a priori”.
To that, we have very simple and reasonable empirical answers: the only biological self-replicators that we know to exist are prokaryotes, and the first prokaryote of which we can empirically know anything is LUCA. And even prokaryotes are not “Darwinian capables”, if not for trivial cases of microevolution.
And no, LUCA was not within reach of neo darwinian processes. But, as for that, neither were all the following species that emerged.
RodW,
It is STILL a matter of EVIDENCE. Also what “evolution” are you referring to?
An interesting fact as to the universal RecA protein, which Dr. Durston mentions here in these videos,,
and which Dr. Durston also mentions here in this paper,,
Adding to the ‘context dependency’ of aa sequences for RecA (and other proteins) which Dr. Durston mentioned in the preceding paper, it is now found that his “universal RecA protein” is also found to be involved in a rather ‘miraculous’ feat of extreme genome repair which takes this ‘context dependency’ issue to a whole other level!
i.e. How can RecA proteins possibly know how to reconstruct ‘shattered’ DNA? If the Genetic/Molecular reductionism model of neo-Darwinism were actually true this should not be possible? Clearly the information for how the DNA is to be reconstructed must reside elsewhere than within the shattered DNA or even within the RecA protein.
RodW:
I am afraid that still your question is not completely clear, but I will assume that it corresponds to my interpretation a) in my post #34. Is that correct?
However, I think I have tried to answer both possible meanings of your question, in that post.
gpuccio
There are plenty of hypotheses that can be tested, and are. In any case, ancient prokaryotes are themselves putative – we don’t have any to observe! They are an inference, from perfectly valid data and data analysis. The same techniques can potentially take us back further.
Obviously you are entitled regard such work as “fantasy” but many scientists beg to differ, and for good reason. I think it needs a more considered rebuttal than mere dismissal as “fantasy”.
Well, as you say, they emerge in populations of organisms, and yes, they have new sequences and functions. My point is that you do not need new proteins for either speciation or adaptation to occur.
“LUCA” is not “empirically known” to be the Origin of Life. It is merely an inferred organism, or population, regarded as the Last Universal Common Ancestor. That inference does not imply that it was also the First. And while you might regard anyone who does not regard the LUCA as the FUCA as not “reasonable”, there are plenty of them. There is simply no reason to make that assumption.
Well, it’s not a “simple fact” – it’s an inference. But I’m certainly not denying it. It supports my case, not yours – that the majority of protein superfamilies emerged after Darwinian-capable organisms are inferred to exist. Therefore Darwinian processes, not random draw, must considered as possible non-design origins for them.
The more interesting problem is how the protein superfamilies that were present at LUCA got there. Most biologists would say they evolved – i.e. assume that there were many many populations ancestral to LUCA, all capable of evolving.
Again, you may disagree, but there’s nothing intrinsically “unreasonable” about this.
Not really. Obviously you can’t invoke something that doesn’t yet exist to account for itself.
You deny that Darwinian processes account for evolution at all? Really?
Well, I think we’d better just leave it there, gpuccio. I didn’t realise that you denied Darwinian evolution completely. I thought you just thought it insufficient.
But I will say that perfectly reasonable people are perfectly capable of being persuaded, by evidence, that Darwinian evolution i.e. adaptation by means of heritable variance in reproductive success in the current environment, not only works, but can be observed to work, in lab, in field, and in silico.
Absolutely, yes. I explained why in my answer to you but it is better to think of alleles instead of proteins and to think of populations instead of an individual protein. A population could have several different versions of the protein, each slightly different and this represents an increase in information.
I haven’t read all the other replies but I doubt they say something different.
Elizabeth:
We can certainly leave it there. Just a few clarifications:
a) I, like many other people, have dedicate much time here to “considered rebuttals” of current OOL “theories”. I dismiss them as fantasy, but I have explained many times why I do that.
b) All science is made of inferences. Some are good, some are bad. But there is nothing else in empirical science. Only inferences to explain facts.
c) I do deny that darwinian processes account for “macroevolution” at all. They can generate no complex functional information. They have a minor role in microevolution and simple adaptations. If, after all the time we have discussed, you have not yet understood that this is my position, there is really no hope.
d) I do believe and maintain that your “perfectly reasonable people” stick, for reasons that are completely mysterious to me, to perfectly unreasonable convictions and ideas. Is that an unpopular belief? Probably. But it’s exactly what I believe.
Mung
The minimum system required for Darwinian evolution is a population of self-replicators that replicate with heritable variance in reproductive success in the current environment.
I do not know – no-one does – what the simplest molecular assembly is that can do this is. If we were to find one (for example, Szostak’s populations of lipid vesicles containing self-replicating double polymers), I don’t know whether you’d call it “alive” or not. It would satisfy some traditional criteria for living things, anyway – growth, reproduction and nutrition.
The virtual organisms in an evolutionary algorithm satisfy some criteria for life, but not others.
Yes.
gpuccio:
OK.
Or models to fit data 🙂 I agree.
I thought it was your position, but your post seemed to contradict that. Thanks for clarifying.
OK.
Actually it is better to think of the information in a gene pools as opposed to specific proteins. I am sure Durston has a much better answer. But the basic answer is yes, information can be increased by some small amounts and happens quite frequently. By the way information decreases from gene pools quite frequently as certain alleles/genes get eliminated.
Dr Liddle, for a replicator to be “Darwinian capable” it must have the capacity to translate a genotype into a phenotype.
You pose the question in a manner to avoid the answer. We do not know “how simple” it can be, but we do know how complex it must be – at least at a minimum, in order to accomplish Darwinian evolution. Those requirements have been explained to you in a coherent physical model which has been verified by both experiment and logic. Exactly none of the key hurdles in that model have been shown to be associated with mere “chemistry and physics”. What is gained by ignoring these facts? To simply repeat “I do not believe the model” is insufficient for empirical discourse. If you cannot demonstrate it to be false, and if you choose not to accept it provisionally, then you are left to simply ignore it. There are no other alternatives.
Elizabeth:
Please tell us the difference between darwinian processes and a random draw.
Well Tracy and Lincoln had a population of self-sustained replicating RNAs that had heritable variation. But nothing new evolved.
Evolutionary algorithms are evidence for intelligent design evolution, not darwinian evolution. Ya see Lizzie, EAs have a goal or goals and darwinian processes do not.
Elizabeth:
Unlike you, perfectly reasonable people understand what darwinain evolution entails. And it is more than just “heritable variance in reproductive success in the current environment“.
Natural selection is cumulative and not random.
Are you sure you’ve really understood evolutionary theory?
Jerad:
Evidence please. Cumulative selection = ID
And BTW, could you please link to this alleged “evolutionary theory” so we can all read what it says. Thanks.
The fossil record, the DNA evidence, bio-geographic distributions, morphology. Cumulative selection is just one round of environmental culling on top of another, on top of another, etc.
You know, I’m sure you’ve heard all the explanations before and so your asking for another round of clarification is just time wasting really. No matter what I say you’re going to claim that my evidence isn’t adequate. BUT you won’t have any good, positive counter evidence and you’ll assert that finding gaps in our current knowledge is enough to call the whole of evolutionary theory into question.
I think there are ID proponents who really are interested in having a dialogue and considering the actual evidence. But, sadly, there are others who just look to score immediate points on blogs which have nothing to do with the real evidence or research.
When can we get past the endless repetition of restating what evolutionary theory does and doesn’t say? And, more importantly, can we please give up the notion that the Uncommon Descent forum is the proving ground? I might argue my case badly but it doesn’t mean my ‘side’ has lost. IF you all really are interested in the best case for evolutionary theory then you won’t find it here, or in Dr Dawkins books, or at any one website. You’ll find it in the accumulated knowledge generated by 150 of work done by biologists probing and testing and expanding on the basic framework. Darwin’s idea survived the insight given by Mendel. It survived, and was upheld, by the discovery of the structure of DNA. Modifications to the basic theory have had to be made as one would expect. All knowledge is provisional. But the core is very solid. And to be asked for evidence now that natural selection is not random is just showing a wilful and intentional ignorance about the model.
If you want to be taken seriously then stop making ignorant arguments. Show some insight and some knowledge and make some real criticisms. Like Dr Margolis. Goodness knows, that woman had to fight hard to get her ideas accepted. But she did because she did the work and had the evidence and really understood the basic theory. And asking questions that have been addressed many, many, many times before is not understanding the basic idea. It’s just wasting time and trying to score points. And I wish some of the more insightful ID proponents would call their fellows on such pointless behaviour.
But you won’t. You all have this conspiracy theory mindset and we’re ‘the enemy’ against who you all must strive. And even bad counter-arguments are good because they uphold the side.
If you don’t know what it says then how do you know you disagree with it?
Upright Biped
Well, it must have heritable variation in reproductive success. The thing that has reproductive success is the phenotype. The thing that makes it heritable is what you might call the genotype, but at its simplest, the genotype is the phenotype. For example if you had a double stranded polymer that tended to split into two single strands (as the Szostak lab has suggested), and those single strands then bonded with corresponding monomers in the environment to produce two versions of the original double strand, you’d have a self-replicator with a genome that was also the entire phenotype. And if, for example some monomers are more prevalent in the environment than others, then those polymers whose segments frequency most closely matched the distribution in the environment would tend to reproduce most successfully. So you’ve got a Darwinian-capable self-replicator already. Szostak also suggests that the combination of such polymers and lipid vesicles that enclose them provides a slightly more complex phenotype, where again, very simple properties of the enclosed polymers, length, for intance – could affect reproductive success – for instance by optimising osmotic pressure gradients.
So while I agree that a phenotype-genotype relationship is important, and that, specifically, the genotype needs to affect the reproductive success of the phenotype (“heritable variance in reproductive success”) to be capable, by definition, of Darwinian evolution, there is no need for that relationship to involve a “translation” process as we now know it. Nor even, once we are talking about RNA, do we need even to involve proteins – RNA is a double polymer capable not only of self-replication but of self-catalysis, offering the possibility, again, of catalysing compounds within the vesicle that could make it more robust, for instance.
I’m not saying these things did happen, just that they are hypotheses currently being investigated, and which do not involve the modern DNA-RNA-Protein pathway and yet would be Darwinian-capable.
I’m not trying to “avoid the answer” – I don’t know the answer. No-one does. That’s why there is active research going on. We don’t know how complex it must be at a minimum. That’s why I said we don’t know how simple it could be. It’s the same question, and we don’t know the answer.
Not in any way that persuades me that the DNA-RNA-protein process is the simplest Darwinian-capable process. You have explained your case, but that doesn’t mean your explanation holds water. I don’t think it does.
Because they aren’t facts, Upright Biped. I know that you think that there is something other than “mere chemistry and physics” going on, but I see no reason to think so. We know that what goes on in the cell requires nothing other than chemistry and physics of the molecules. They are certainly very fancy molecules, and I understand that IDers think they can only have come into existence by means of some kind of intelligent agent, but they are, nonetheless, biochemical molecules and they do what they do because of physical and chemical forces in the cell, not anything else.
And nothing you have said persuades me that there is some necessary barrier between a Darwinian-capable pre-DNA system and a DNA system. Indeed there is a body of research on the evolution of the ribosome, for instance.
There might be a barrier, but I see no reason to infer one, just because the DNA-RNA-ribosome pathway is the way it is.
Jerad you state:
And exactly what does cumulative Natural Selection accumulate?
Hmm it’s not proteins that it accumulates,, In fact the overwhelming tendency of Natural Selection is to eliminate (purify) a genome of non-functional elements,,
To the extent Natural Selection can be said to accumulate anything Jerad, it can be said that Natural Selection is excellent in accumulating slightly detrimental mutations long before any rare island of functionality will ever be stumbled upon:
the evidence for the detrimental nature of mutations in humans is overwhelming for scientists have already cited over 100,000 mutational disorders.
Also of note;
Jerad, in spite of this consistent evidence, of which I’m sure you are well aware, you still believe in evolutionary theory, to which I can only ask, ‘Are you sure you’ve really understood evolutionary theory?’
Hence significance of 500 – 1,000 bit thresholds as a barrier to blind search for novel islands of function, starting with the first. Where discussing incrementalism within an island begs the pivotal question, finding islands. And how do we know, isolated islands? Multipart function depending on specific arrangement, interfacing and coupling, in a world where otherwise, ever so many arrangements are possible. KF
‘And exactly what does cumulative Natural Selection accumulate?’
First of all, a blueprint, Philip, a design? Then it’s all random systems go! Like working out a rubik cube by accidental fiddle-faddling. It’ll all fall into place just as per the blueprint. Apparently, there’s a strange, little-understood affinity between designs and randomness.
Well we don’t know if we disagree with something we have never seen. We do disagree with several basic conclusions of modern evolutionary theory based on the evidence we have seen. But maybe there is something we are not aware of. That is what the comment about “alleged” evolutionary theory was referring to. Is there something we are not aware of.
You are presenting the “overwhelming evidence” argument and we have seen that before but science does not work that way. A lot of what you present does not indicate any specific mechanism for evolution. So I suggest that you present the best case for how major changes happened. One of the necessities for major changes is the origin of new proteins which is what this OP is about. New proteins are not the only thing necessary but it is one of them. What determines where these proteins go and in what sequence is unknown as of this moment so is another major issue. New proteins however, present a very thorny problem.
From your #11, JLAfan2011:
‘Show me on natural thing by experimant in the lab or in the field that a designer was invloved in. I don’t want human made examples and I don’t want evidence after the fact.’
I can’t afford to visit a beaver-lodge construction site on my own, never mind organise a field trip for you, fer goodness sake.
Though I did think disqualifying human beings as designers was a bit rich.
Yup. The mutual support given by multiple lines of independent evidence. over 150 years of research and data collection, the fact that the theory has withstood various and sundry criticisms and attempts to knock it down. In fact, some of the strongest evidence in support of the theory has come in the last 60 years with the discovery of the structure of the DNA molecule and the genetic evidence of transposons, ERVs and the coding redundancies. For any other phenomena the incredible volume of confirmatory evidence would be more than enough.
Still I’ve thought long and hard about it. And, I admit, it seems fantastical at times. But there is no better explanatory model. There’s no hypothesis that comes close to lining up with ALL the data. It’s elegant and it works. And it requires no assumption of some undefined, undiscovered agent.
It has predictive power, it lines up with all the data, it’s supported by multiple lines of evidence, we are even starting to see it ‘in action’. I’m just glad I was born at a time when we are figuring out how we came to be. Its fascinating. It helps me to appreciate the wonders of nature and how lucky we are to be around at a time when the structures beneath the surface are becoming discernible.
Nature is very good at trial and error. I’ve read that something like a third of all human conceptions end in spontaneous abortions. Many before the woman is even aware that she’s pregnant.
Lots and lots of mistakes are made and those attempts are not fit, they don’t survive long enough to reproduce. We are the lucky ones. Our mix of genes was good enough.
IF you really are interested in all this stuff then I suggest you read current research papers and not just books written for the general public. I enjoy Dr Dawkins’ books but they are not where the action is. That’s in the research and journals being published every day. Little bits and pieces adding to the overall picture which hasn’t substantially changed since 1859. The gaps are decreasing as more and more of the process becomes clear. It’s a very cool time to be alive, to see the ‘secrets’ revealed.
You want to cast doubt on the whole thing just because you can’t conceive of how some steps occurred. Asking questions is fine but being a merchant of doubt eventually becomes a idealistic stance. Just like the tobacco companies that tred to drag out the doubt that smoking is bad for you. Or the people that said the evidence wasn’t clear that CFCs were harming the ozone layer. Or the current crop of climate change doubters. Sixty or seventy years ago many people couldn’t accept that the continents move.. Even now parts of the mechanism are not fully understood. But the continents do move and few intelligent people would deny it.
I think it’s easier to buy into the ID hypothesis than to wrap your head around evolutionary theory. I think most of us aren’t wired to accept the fact that a combination of random mutations and eons of cumulative selection resulted in us. Who would want to think that? But that’s what the evidence points to. And the data is growing. I didn’t naturally come to believe in evolutionary theory; I had to be convinced. And I ran out of other explanations that encompassed all the data and had less special pleading.
Jerad, I couldn’t help to notice in all your romantic reflection about your Darwinian worldview, and all the lines of evidence you alluded to supporting it for the past 150 years, that you forgot to actually cite an actual example of Darwinian processes creating anything. I know in love affairs, such as you have for Darwin, it is hard to see faults in the one you love, but, as a friend, let me let you in on a few defects that you have overlooked:
And in spite of the fact of finding molecular motors, and highly sophisticated control systems, permeating the simplest of bacterial life (Alberts, Venter), there are no detailed Darwinian accounts for the evolution of even one such motor or system.
The following expert doesn’t even hide his very unscientific preconceived philosophical bias against intelligent design,,,
Yet at the same time the same expert readily admits that neo-Darwinism has ZERO evidence for Darwinian processes producing any cellular system whatsoever,,,
Though Michael Behe has been pointing this failing in Darwinism out for decades, the only refutation offered by Darwinists has been a very deceptive literature bluffing:
Morever, despite the fact that Darwinists have ZERO empirical evidence of Darwinian processes EVER producing any molecular machine, here is an example that intelligence can accomplish as such:
The same situation is found with proteins. i.e. Even though no one has ever observed Darwinian processes creating a single functional protein from scratch, intelligence has demonstrated the capacity to create as such:
So Jerad, seeing such stark disparity in empirical evidence compared with your unyielding infatuation for Darwin, I can only ask once again, ‘Are you sure you’ve really understood evolutionary theory?’
Dr Liddle,
Do differences in material processes matter, and do you think it is important to acknowledge those differences? What exactly is your standard of discipline with regard to identifying individual material processes and assigning the capacities of one process to the observation of an unrelated process? And if, for some reason, you should suggest that one process can lead to a target outcome, and then find that the proposed processes cannot (in fact) result in that target, is it your position that you should cease from making that proposition in the future, or will you continue to do so?
Elizabeth Liddle:
Darwinism. A system of belief in search of a theoretical underpinning.
So here we have it, in plan text, for all to see, that Elizabeth admits she was just pontificating with no real theory or evidence to support her assertion.
Par for the course.
Elizabeth Liddle:
The logic is impeccable.
Why, Elizabeth, can’t one invoke Darwinian processes to account for the first Darwinian-capable self-replicator?
Perhaps there is a “pre-life” Darwinian process and a “post-life” Darwinian process.
What makes a replicator “Darwinian-capable,” whatever that means?
Well, you really can’t say, can you, since you don’t even know what the simplest system capable of Darwinian evolution must consist of. In fact, you don’t even have a theory about what such a system must consist of.
And you really don’t know WHEN you can invoke “Darwinism” as an explanation, as you just admitted as outlined in my post @71.
Hand-waving is not, and never has been, science.
You have resorted to an ad hominen attack. That is the first sign that one does not have anything to back up their position.
I suggest you cite a few and we can evaluate them.
Also what do the tobacco companies or global warming or CFC’s or plate tectonics have to do with evolution? Keep to the topic at hand, in this case information in the gene pool. If you can expose us as ignorant, go for it. Millions will be ecstatic if you could do that.
Shame on you gpuccio. Stuff just happens, that’s all, is recognized by all to be an indisputable truth and a more than adequate explanation in the hallowed halls of science.
Jerad:
Culling is the exact opposite of accumulating.
Jerad:
This is truly rich coming from you Jerad.
Invitations were repeatedly extended to you, yet you declines each one. Go figure.
Mung
It is indeed, Mung. Re-read it carefully if you are not truly convinced.
Jerad:
I’ll tell you when. When you and other advocates in favor of Darwinian evolutionary theory can present to us a testable theory for the simplest system capable of Darwinian evolution.
If you don’t even know what is required for Darwinian evolution, and have no theory capable of being tested in support of the claim of what is required, you have nothing scientific to present to us.
Is the minimal system capable of Darwinian evolution capable of recognizing sarcasm?
Jerad:
And how is any of that evidence for natural selection?
Look Jerad, Dawkins modeled cumulative selection with his weasel program. It ain’t darwinian.
You can’t even muster testable hypotheses complete with predictions wrt unguided evolution. And Lenski’s long-running experiment exposes your position’s claims as total bogus.
So perhaps you should pull your head out, take a look around and see what reality really says.
Jerad:
One college textbook on Macro-evolutionary theory Jerad, just one. I’ll buy it and read it. But as yet I can’t find it.
Maybe there’s no money to be made in publishing textbooks on macro-evolutionary theory. yeah, that’s must be it.
I bet you had math textbooks in uni, right? Why should macro-evolutionary theory be exempted?
Jerad:
Yes it is easier to buy into reality than wrap your head around something that doesn’t exist.
Nice call.
Mung, in case you missed it-> If you want a book on macrevolution then all you do is get a bunch of nooks on microevolution and put them together. After all macroevolution is just microevolution, microevolution and more microevolution. 😉
Jerad:
I know what evolutionism says, Jerad. I also know that it isn’t a theory. That is because it can’t muster testable hypotheses along with predictions.
Hi Elizabeth,
Sorry I missed you during the period when I was active at your blog. We would have had so much to discuss.
I’ll admit that I find it challenging to take you seriously, but if you’ll try hard I’ll try hard.
You claim that one cannot invoke darwinian evolution to explain the existence of biological systems capable of darwinian evolution. You seem to think this claim is imminently logical. Yet the facts clearly contradict your claim.
Darwinian evolution is invoked in all sorts of situations that have nothing to do with biological organisms.
And there’s no logical reason that one “Darwinian process” cannot give rise to some other “Darwinian process.” Do you disagree?
Jerad:
What makes you think this is a “trial and error” process?
Jerad:
What theory?
Numerous turtles hatch and head for the sea. A bird swoops in and snatches one and makes a meal of it. It fails to survive and reproduce. It’s “less fit” than it’s brothers and sisters, by definition.
What qualifies that as a theory? How is it testable? What makes such a theory scientific?
Mung:
Mung:
Elizabeth Liddle:
’nuff said
Care to claim that I misrepresented what you wrote Dr. Liddle? I don’t think I did.
So do you grant my point?
Axel wrote:
“I did think disqualifying human beings as designers was a bit rich.”
Dembski calls human beings ‘mundane designers’ to distinguish them/us from ‘transcendent designers’ (2004). So, it’s not like he ‘disqualifies’ them/us, but that he underplays the significance of actually studying ‘human designers,’ which indeed would undermine the paradoxically both vast and narrow claims of ‘Intelligent Design Theory’ qua ‘theory.’ Such a situation explains the peculiar unanthropic bias/composition of DI-CSC fellows.
This is why the Big-ID vs. small-id distinction is so important and also why several theists who reject IDism have pointed it out…and been ignored or scorned by IDists. Durston violates the distinction – opting to try to have his cake and eat it too. That is, does not allow the distinction to be expressed or hasn’t yet addressed it, just like some other IDists who would try to universalise the single concept ‘design’ into a kind of ‘designism’. That is his bias, which his linked ‘Executive Summary’ on http://www.powertochange.com displays quite clearly. That’s what I call a bait-and-switch, and it is rampant in the IDM, including here at UD.
It also explains why IDist leaders of the IDM at the Discovery Institute, though Durston is not a DI Fellow, have not taken much time to study ‘design theory’ as it is elaborated in fields outside of biology, cosmology, geology, informatics and engineering. Even the coiner of ‘Intelligent Design’ Charles Thaxton’s ‘borrowing’ (perhaps call it faulty concept transfer) of the term ‘intelligent design’ from engineering betrays a lack of explicit explanation of how ‘mundane designers’ and ‘transcendent Designer(s)’ differ or are similar and on what grounds he compares them.
And then there’s Meyer’s notion of ‘historical sciences’ (“Causes now in operation,” – Lyell, i.e. “causes that are known to produce the effect in question” – Meyer) as simple ‘reverse engineering/reasoning’ that falls flat when it is understood that no human being in history has ‘experience’ of creating or designing the universe we humans live in. OoL is not a topic/field/theme that Meyer’s ‘historical science’ approach can reasonably apply too; though invoking mystical origins/Origins is certainly possible (if not ‘scientific’).
Is ‘Origins of Life’ a ‘historical (or forensic) science’? No, of course not. No one we know (other than perhaps [a] transcendent Designer[s]) was there/then.
Durston’s ‘Executive Summary’ is simply another example of an IDist, promoting his/her blatantly scientistic approach to IDT, on an evangelical Christian website, as if the scientific community should pay attention and ‘revolutionise’ itself to become something that natural science is not. Unfortunately for Durston, too many people have already seen through the IDM’s smokescreens and self-contortions, to its ‘little-Big-tent’ watering-down and catering to YEC funding channels. The life of the movement qua movement, outside of creationist/evangelical Protestants in the USA (and a few Canadians, like Durston, Bruce Gordon [NYC], Denyse, etc., is already gradually fading away due to honest, open and also faithful scientists and scholars rejecting IDism as a pipe-dream of questionable political PR credibility, the BioLogic Institute (and the collapse of ISCID.org, failure at Dover, cdesignproponentsists, Dembski’s ‘Waterloo’ at Baylor, etc.) included. But there are those who think martyrdom is an IDist virtue worth displaying on forums like UD, ASA, etc., so it would be hard if not impossible to change such persons’ hearts to ever accept this ‘interpretation’ of facts and evidences.
The bottom-line is that responsible and devout Muslims, Christians, Jews and Baha’is should openly and directly reject ‘Intelligent Design Theory’ qua ‘theory’ as 1) unecessarily scientistic ideology, 2) unfaithful to the traditional Abrahamic religions that claim no ‘scientific’ proof (they soften/weaken, try to sweeten it to be simply an ‘inference,’ but we’ve seen their leaders’ hearts when speaking/writing about Big-ID in religious environments) of the Creator’s existence was/is/ever will be possible, and 3) an approach based on an amazingly and suprisingly (for America, which is supposed to be so ‘scientifically advanced’!) weak philosophy of science (most IDists don’t look beyond Thomas Kuhn), which depends far too much and inexcusibly on statisticism and probabilism (don’t just quote dictionaries, think forward, bigger and better than IDism!) that are ideologies the IDM will try to fight against though they are appropriate labels of its ‘attitude’ towards knowledge until it soon becomes extinct in the history of ideas.
But hey, Durston’s ‘scientific case’ (regurgitated from others) might end up being a good advertising strategy for those who already believe in the ‘power to change’ via a transcendent Designer, who think as he does that de-capitalising ‘intelligent design’ will deceive or trick the majority of religious folks to trust the Discovery Institute and their political ‘Wedge’ mission of desperately attacking the ghost of ‘Darwinism’ past. Those people, already religious believers in mainly Protestant (with a trickle of Catholic) funding channels, is clearly what the IDM has ‘banked’ on so far.
Please return back to your “nf/nt ratio approaching zero” now and ignore this message. Probably.
for instance
You seem to be vying with JoeG for culture-warrior-par-excellence. I think you are in with a good chance if you keep up with comments like these.
Think about this for a moment, if you can. The ToE is usually summarized and starts with the assumption of an existing population of self-sustaining self-replicators. Nobody yet has an evidence-based theory for the origin of life on Earth (mainly due to a dearth of evidence, not a dearth of theories). Sure evolutionary processes (trial-and-error)is a very effective approach in many fields.
What point are you trying to make? Are you trying to obscure the lack of a theory of ID (evidence-based or not)? I don’t see Dr. Durston making much progress here, either. Same old default to design inference with no idea what a design inference entails.
Mr. Fox, as to “with the assumption of an existing population of self-sustaining self-replicators”
Even giving atheists the ‘assumption’ of protein making machines that are made out of, you guessed it, proteins, do you have any evidence that Darwinian processes can make, you guessed it again, proteins???
Notes:
In fact the Ribosome, which makes the myriad of different, yet specific, types of proteins found in life, is found to be severely intolerant to any random mutations occurring to proteins.
The Ribosome: Perfectionist Protein-maker Trashes Errors
Excerpt: The enzyme machine that translates a cell’s DNA code into the proteins of life is nothing if not an editorial perfectionist…the ribosome exerts far tighter quality control than anyone ever suspected over its precious protein products… To their further surprise, the ribosome lets go of error-laden proteins 10,000 times faster than it would normally release error-free proteins, a rate of destruction that Green says is “shocking” and reveals just how much of a stickler the ribosome is about high-fidelity protein synthesis.
http://www.sciencedaily.com/re.....134529.htm
And exactly how is the evolution new life forms suppose to ‘randomly’ occur if it is prevented from ‘randomly’ occurring to the proteins in the first place?
As well, the ‘protein factory’ of the ribosome, which is the only known machine in the universe capable of making proteins of any significant length, is far more complicated than first thought:
Honors to Researchers Who Probed Atomic Structure of Ribosomes – Robert F. Service
Excerpt: “The ribosome’s dance, however, is more like a grand ballet, with dozens of ribosomal proteins and subunits pirouetting with every step while other key biomolecules leap in, carrying other dancers needed to complete the act.”
http://creationsafaris.com/cre.....#20091010a
As well, The Ribosome of the cell is found to be very similar to a CPU in a electronic computer:
Dichotomy in the definition of prescriptive information suggests both prescribed data and prescribed algorithms: biosemiotics applications in genomic systems – 2012
David J D’Onofrio1*, David L Abel2* and Donald E Johnson3
Excerpt: The DNA polynucleotide molecule consists of a linear sequence of nucleotides, each representing a biological placeholder of adenine (A), cytosine (C), thymine (T) and guanine (G). This quaternary system is analogous to the base two binary scheme native to computational systems. As such, the polynucleotide sequence represents the lowest level of coded information expressed as a form of machine code. Since machine code (and/or micro code) is the lowest form of compiled computer programs, it represents the most primitive level of programming language.,,,
An operational analysis of the ribosome has revealed that this molecular machine with all of its parts follows an order of operations to produce a protein product. This order of operations has been detailed in a step-by-step process that has been observed to be self-executable. The ribosome operation has been proposed to be algorithmic (Ralgorithm) because it has been shown to contain a step-by-step process flow allowing for decision control, iterative branching and halting capability. The R-algorithm contains logical structures of linear sequencing, branch and conditional control. All of these features at a minimum meet the definition of an algorithm and when combined with the data from the mRNA, satisfy the rule that Algorithm = data + control. Remembering that mere constraints cannot serve as bona fide formal controls, we therefore conclude that the ribosome is a physical instantiation of an algorithm.,,,
The correlation between linguistic properties examined and implemented using Automata theory give us a formalistic tool to study the language and grammar of biological systems in a similar manner to how we study computational cybernetic systems. These examples define a dichotomy in the definition of Prescriptive Information. We therefore suggest that the term Prescriptive Information (PI) be subdivided into two categories: 1) Prescriptive data and 2) Prescribed (executing) algorithm.
It is interesting to note that the CPU of an electronic computer is an instance of a prescriptive algorithm instantiated into an electronic circuit, whereas the software under execution is read and processed by the CPU to prescribe the program’s desired output. Both hardware and software are prescriptive.
http://www.tbiomed.com/content.....82-9-8.pdf
LIFE: WHAT A CONCEPT!
Excerpt: The ribosome,,,, it’s the most complicated thing that is present in all organisms.,,, you find that almost the only thing that’s in common across all organisms is the ribosome.,,, So the question is, how did that thing come to be? And if I were to be an intelligent design defender, that’s what I would focus on; how did the ribosome come to be?
George Church
http://www.edge.org/documents/.....index.html
Of note, although the ribosome is present in all life, it is not uniform across all life:
Ribosome
Excerpt: Ribosomes from bacteria, archaea and eukaryotes (the three domains of life on Earth) differ in their size, sequence, structure, and the ratio of protein to RNA.
http://en.wikipedia.org/wiki/Ribosome
But even spotting all that to the atheist, does the atheist have any evidence for proteins can evolve? No!
Doug Axe PhD. on the Rarity and ‘non-Evolvability’ of Functional Proteins – video (notes in video description)
http://www.metacafe.com/watch/9243592/
Yes, Mr Cunningham?
That’s not the issue. ToE proposes a method of reiterated variation and selection that can result in phenotypic change over time. I don’t ask you to believe it, just try not to misrepresent it.
Alan Fox is lying as there isn’t any ToE. He cannot link to it and he cannot reference it.
Not only that Alan is still ignorant wrt the meaning of “default”.
And finally only the willfully ignorant cannot grasp that how life originated directly impacts how it evolved.
So Mr. Fox, you are basically saying that even if we grant you the ‘assumption’ of protein making machines that are made out of, you guessed it, proteins, you still do not have any evidence that Darwinian processes can make, you guessed it again, proteins??? And the point of avoiding the OOL issue was what exactly?
Top Ten Most Cited Chemist in the World Will Buy You Lunch If You Can Explain To Him How Evolution Works – James Tour, Phd. – video
http://www.youtube.com/watch?v=WCyAOCesHv0
Actually, I buy that as the underlying theory that a very large number who are pro naturalistic evolution actually believe. It seems to be what several are defending here. My problem with this theory which can be demonstrated to actually work in the real world is that the only phenotypic changes that can be pointed to are trivial. It does not lead to anything of consequence in the evolution discussion.
So I believe this is an accurate description, but an irrelevant one to the total discussion. Is there anyone who is pro naturalistic evolution who disagrees with this assessment or wants to amend this statement in any way? Because if there isn’t then we can save a lot of pixels and focus on the theory under contention.
The interesting thing is that after 8 years of this blog, there is a discussion about such a simple matter.
Well, Gregory, when it comes down to it, while I hardly understand a word of the technical guff you wrote in #89, re ID, I’m a great believer in the dictum of Watergate’s wily, old, hick, country lawyer, Sam Irving; viz if it looks like a duck, quacks like a duck, waddles like a duck … well supposing instead of that handful of characteristics, we consider in the same light, the punctilious coincidence of 140 odd physical calibrations, required for the existence of our universe, would that not suggest design yet more powerfully yet, even very much more powerfully, than observable features of the behaviour of a duck?
Is inference a sly evasion or trick in such a context? Requiring a mind-boggling conceptual leap? I don’t think so. I don’t know about you, Gregory, but I reckon a lot of your anti-ID confreres must need their mothers to do up their shoe-laces in the morning.
‘I know what evolutionism says, Jerad. I also know that it isn’t a theory. That is because it can’t muster testable hypotheses along with predictions.’ – Joe
Yeah, but there could be a multiverse in which it can, Joe!
Darwinism, multiverses…. You know what? I think the desperation they indicate, must be a function of a desperation verging on outright despair, on the part of atheists, for an emotional crutch.
Only an overpowering emotional craving could deprive them of their wits in such an overt, obdurate and scary fashion.
Elizabeth Liddle:
Gerald Joyce Defines Life as Darwinian Evolution
Oh sure, OOL and Darwinism are completely separate.
But what theory defines what constitutes a system capable of Darwinian evolution?
No theory, not science.
Having spent a short and most enjoyable vacation up on St. Josephs Island, Lake Huron, I now see there are a considerable number of comments. I cannot respond to everything, but will do my best to cover what I can by comment # and user ID.
#26 Gregory: It appears to me that you may be conflating two issues: the scientific issue of whether an intelligent agent was involved in the origin and diversification of life, and the philosophical issue of who or what that agent might be and the implications thereof. Design detection is already firmly entrenched in science in areas such as forensic science, archeology, SETI. Whether we like it or not, it now needs to be applied to biology with the advent of artificial proteins and genomes. The job of science is to develop a general method to detect cases where intelligence was involved. The job of philosophers is to deal with the implications. A critical thinker will recognize the distinction. Unfortunately, some scientists see the philosophical implications of biological life testing positive for design and suspect that the implications point to God. The result is that they fail to do the science because of the philosophical implications. Worse, they become inconsistent in their practice of science, recognizing that design detection is a firmly entrenched branch of science but making an arbitrary, philosophy-driven exception when applied to biological life, in which case design detection is not science but ‘religion’. That is a flat-out failure to think critically and distinguish between the job of science and the job of philosophers. The job of science is to develop a scientific method for design detection, apply it to biological life and leave the philosophical implications up to the philosophers and theologians. If you believe that if the science of design detection indicates that biological life tests positive for intelligent design, and God would be the best candidate, that is fine, but the philosophical/theological implications should not scare one away from doing the science.
#89 Gregory: I am puzzled by your belief that my executive summary is biased, or some sort of bait and switch. I regard it as cold, hard, testable science that makes predictions and leads to further research. You will need to show me which line or lines in the executive summary is the ‘bait and switch’ or which line is biased. Good scholarship requires that one accurately represents the other person’s position (which you have failed to do) and defend your analysis with specific examples (which you have also failed to do). So let us get some detail into this conversation … which line in the executive summary do you have a problem with? I do hope that you would not be willing to throw science out the window due to your fear that the science might actually point to a designer of life (i.e., God). The whole case laid out in my executive summary depends upon a scientifically testable and falsifiable hypothesis. Everything else follows from that hypothesis. All you have to do is falsify the hypothesis, so have at it.
#30 RodW: Your question of whether small, randomly produced bits of functional information (FI) can accumulate to eventually produce a stable, functional 3D protein structure is an excellent hypothesis that can be tested scientifically. First, it is possible to ratchet up the FI of any sequence or system provided a) each step is trivial enough to be accomplished by a random search and b) there is a large enough fitness advantage to preserve each step. We can demonstrate this with GE’s (genetic algorithms), which are examples of intelligent design in action provided some intelligent thought is exercised in designing the fitness function within the algorithm. What we have learned from GE’s, however, is that the fitness function must be carefully designed to guide the search toward some objective. When it comes to the tens of thousands of protein families, there are not tens of thousands of carefully crafted fitness functions to guide the evolutionary search. All we have is one very crude fitness function which we call natural selection, which according to recent work may be useful in fine-tuning actual genomes, but work against straying too far from the fitness space of existing organisms and protein families. Secondly, my work, and the work of many others, reveals that the hypothetical evolutionary steps to build a novel protein family would have to be huge, at least on the level of structural domains. To put it more simply, the steps would have to be quantum jumps not small steps as Dawkins envisioned in his ‘climbing Mount Improbable’. To put it another way, most of sequence space codes for non-stable, non-functional proteins. The average distance between the stable folding islands within sequence space is huge, even for structural domains.
#38 Elizabeth B Liddle: On the basis of our knowledge of the laws of physics, which produce repeatable effects, given identical initial conditions, we can conclude with confidence that natural processes cannot explain the OOL for the following reason. If a result is tightly constrained by the laws of physics (e.g., a crystal lattice), then it will be highly ordered. We can call this ordered sequence complexity (OSC) which is not capable of carrying a statistically significant level of FI. If the result is not tightly constrained by the laws of physics, then the result tends to be what we can ‘random’. We can call this random sequence complexity (RSC). Information carrying sequences contain functional sequence complexity (FSC) and are observed only in computer code, human languages, and genetic information encoded in the genomes of life. Regardless of whatever natural process one wishes to consider, sooner or later it will have to produce FSC if we are to get life and natural processes, which produce only a mix of OSC and RSC. Natural processes are incapable of producing an statistically significant level of FSC for reasons published by Abel and Trevors (1). In that paper they state that ‘rational agency’ is required.
Regarding speciation, etc. if speciation requires only a trivial change in FI, then it is possible. The real issue is how we classify life. The recent discovery of orphan genes, which suggest that different kinds of life require large, quantum changes by way of the addition of novel protein families, makes it necessary to rethink the way we classify life forms. I would say, at the very least, that if a taxon requires one or more novel protein families, then it requires a statistically significant level of novel FI. Since the ability to generate statistically significant levels of FI is unique to intelligence, we have a way to put together a new way of classifying life according to design, that is much more consistent with recent advances in genetics.
With regard to your statement that ‘new proteins simply emerge’, my question is what kind and by what mechanism? When we look at the values of FSC I published in the paper I cited in my original post, and solve for the probabilities embedded in the equation published by Hazen et al (also cited) we see that novel, functional, folding protein families have such a low probability (or low frequency of occurrence in sequence space for those who do not like talking about probabilities), that we cannot expect any functional protein families to be discovered anywhere in the universe, over its entire life span, using an evolutionary search. On the other hand, the level of FSC they require is easily generated by intelligent human beings. Indeed, we are now able to build simple artificial protein structures, another example of intelligent design in action in the lab. On the other hand, if all we need is a random sequence, a randomly generated start and stop codon and the copying and pasting of a promoter region, then it is entirely possible that non-functional, de novo proteins can be produced by mutation, insertion and deletion. Indeed, I would predict on this basis that non-functional proteins can be produced by the simple observation that the novel FI required to produce a novel, non-functional protein is trivial or even zero. Any random process can produce non-functional effects, including non-functional proteins because they are easy to ‘find’ in an evolutionary search. Interestingly, recent work suggests that some de novo proteins are harmful, which is precisely what one would predict for proteins produced by an evolutionary search.
1. Abel DL, Trevors JT (2005) Three subsets of sequence complexity and their relevance to biopolymeric information. Theor Biol Med Model 2:29
KD:
Yes, but you have to understand who you are dealing with here.
For Elizabeth this is no challenge to Darwinian evolution for two reasons:
1.) because it’s “an OOL problem.”
2.) because Darwinian evolution just is just like a computer program and is thus capable of generating FSC.
Elizabeth even had the creative ability to intelligently design a coin-tossing program that could be biased to produce a binary sequence that she believed was beyond the UPB, by which she convinced hereself that Darwinian process could do likewise (or some such nonsense).
It is unfortunate that the other individual to catch your eye was Gregory, a long-time troll here at UD with nothing of interest to offer to anyone. You would do well not to use the term “intelligent design” in any of your posts and of you do, God forbid that you should use capital letters when doing so. That’s the sum and substance of his “contributions” here at UD. Don’t expect him to discuss actual science. That’s outside his field.
“It is, therefore, a mistake to assume that FI can only be generated by intelligence; a trivial level of FI can be produced by completely mindless processes, as should be obvious from the above equation, and as the Georgia Tech results illustrate.”
My problem with this is not only how get all of the chemicals for nt in one place in the universe to begin the so called random sequences, but also the arrangement of sequences (the order) itself seems to have naturalistic assumptions tied to it when applying statistical possibility to produce functional information. Am I missing something regarding the absence of assumption?
“It should be clear from the above equation that if nf is large enough for examples to be generated by random recombinations, then functional information (FI) can be generated by random natural processes, albeit a trivial level.”
How do we know these recombinations are ever truly generated at random? Please be specific. How do we know there is not a requirement of order that would make appeals to a relationship to a total number of possibilities (nt) completely superfluous? (trying to understand your basic assumptions which would allow the mathematical application)
I understand you are ID, KD…but is there a point where natural processes are not really random processes YET naturalists call these processes random processes because they ignore an observable order (or they don’t fully identify/understand the order yet so their accusation of “random” is historically premature)? I would argue against crystalization being random just as I would argue against PNA being formed from random processes. I do fully understand the practicality of speaking the materialistic naturalists’ language…
but I am still here to question basic assumptions.
At last the penny seems to have dropped! As I said in comment # 90:
Even UD currently has a thread on alkaline thermal vents with some more ideas on a possible scenario for OOL.
Aspects of the evolutionary process can be modelled. Kirk Durston is using computer modelling too. It is a bit simplistic to think that means ToE is ‘like’ a computer program. It’s ID proponents that seem to have the idea that cells are ‘like’ nano-machines rather than biochemical in function.
Dr Durston:
But you have no justification for asserting how rare functional (in context) proteins are. The only reliable way of identifying the functionality of a novel protein is to synthesize it and test its properties such as binding affinities. Until we have a short-cut method of predicting functionality in novel proteins, we can only say we don’t know how rare functional proteins are. Though work that has been done (Keefe and Szostak come to mind and there is more recent work) do not suggest this “needle-in-a-haystack” rarity.
Off topic, but this may be of interest to you guys > http://sandwalk.blogspot.com/2.....ttick.html
Excerpt: Mattick also uses IDiot terminology when he says that, “…the argument posits that the presence of non-protein-coding or so-called ‘junk DNA’ that comprises >90% of the human genome is evidence for the accumulation of evolutionary debris by blind Darwinian evolution.” As you should all know by now, the accumulation of junk DNA is the antithesis of “Darwinian evolution.” You should also note that it’s mostly IDiots who get confused about the difference between junk DNA and “non-protein-coding” DNA.
I find that very troubling.
Thanks for your response, Dr Durston.
Let me respond in turn
Yes, I see that. I don’t buy the argument. The paper seems to make the assumption that there is no continuum between slightly functional information (just enough conserved information to count as “replication”) and a fully functioning DNA system with protein-coding and all. There may not be such a continuum, but the paper certainly does not present any argument that there could not be, and as you will know, the focus of OoL labs like Szostak’s is precisely on testing hypotheses regarding such a continuum, many of which seem to be looking very promising. Sure, it requires “choice” rather than “chance”, but that’s exactly what selection is. The issue is just how simple the first self-replicator had to be that had heritable phenotypic success. I’m not saying it’s been found, or even that it ever will be. But to say that the Abel and Trevors paper established for all time that “natural processes are incapable of producing an statistically significant level of FSC for reasons” is over-reaching by quite a margin.
This seems to be a very archaic approach to taxonomy. Speciation, as normally considered in biology, is not an event, but a process, and it’s a population-level process, not a genomic one. At it’s simplest it can result from no more than physical separation, and thus independent drift – not even adaption to different niches. Random walk will ensure that the lineages will diverge if they cannot exchange genetic material. Or are you talking about longitudinal change perhaps? In which case would “macroevolution” perhaps be a better term? Which can, of course, occur without speciation. The problem it seems to me with your argument here is that you are attempting to categorise a continuum. And thus seeing barriers where there are none.
Using evolutionary search? I don’t think so! Only if you assume that there can be no reproductively advantageous precursors, and how on earth do you calculate that?
By “evolutionary search” you appear to mean “random mutation”. Have you forgotten about selection?
Evos still choose willfful ignorance- If te OoL = designed then there is no reason to infer all mutations are genetic accidents which means natural selection is a non-starter.
What selection? Nature doesn’t select and natural selection eliminates, it doesn’t select.
And again it is very telling that neither Lizzie nor Alan can produce any evidence to support their position.
Alan chokes:
Yes, we do, Alan. MIT’s Robert Sauer did the work and it is in peer-review.
OTOH your position still has nothing that demonstrates blind and undirected chemical processes can produce a functioning polypeptide, ie a protein.
Dr Liddle you say;
So is Selection choice? or chance? Since when does non-intelligence make choices? You are trying to tell us that natural selection has some sort of “consciousness” that has the ability to decide what it should or should not select for?
Are you serious?
Are you serious?
Are you serious?
Yes it’s that serious!
If “evolutionary search” necessarily involves selection, then change from near-neutral mutation isn’t considered evolutionary, which would be odd. Since the evidence is increasingly that most genetic changes are not subject to selection (other than purifying) that leaves us with pretty much an evolution-free situation in the genome – not, I’m sure, what Elizabeth intended. If selection is involved, it can only select what has already been sampled by mutation, as Kirk implies, so the maths doesn’t change that much and still depends completely on the shape of the search space.
Anyway, if Nei’s version of neutral theory is right, we shouldn’t lean on selection too much even for phenotypic change.
There are some very good questions raised since I posted yesterday. The Origin Of Life (OOL) is a massive problem and many scientists are careful to distinguish between the OOL and the neo-Darwinian evolution of biological life. There is a general feeling that once the OOL occurred (by some miracle), the diversification of life would inevitably proceed via a neo-Darwinian process.
I contest that assumption; there is another large problem which I will call the Origin Of Protein Families problem (OOPF). The OOL requires only a minimum genome consisting of anywhere from 150 to 460 protein coding genes; the diversity of life we see today requires tens of thousands of protein families. Since the coding for each protein is embedded within the genomes of life in quantized packages we call genes (perhaps simplistically given what is emerging re. the RNA Spliceosome), the OOPF problem is ultimately an origin of FI problem. Our observations of protein coding genes strongly indicate that the FI is quantized, not a continuum.
#101 Breckmin: Your concerns about how we get the chemicals together for the formation of the first life form is certainly valid. For the sake of this discussion, however, I am assuming the existence of at least the Last Universal Common Ancestor (LUCA). This grants the existence of a self-replicating system that can, in theory, evolve (devolve?) through mutations, insertions, deletions, genetic drift, crossover, etc.
Regarding your request for an example where a trivial level of FI has actually been generated, that is difficult though in theory I don’t see why it could not happen. Indeed, it seems inevitable that trivial levels of FI are generated that represent low-level anomalies within the background ‘noise’ of random processes. The only possible example I can think of off the top of my head occurred in Lenski’s Long Term Evolutionary Experiment (LTEE) with the Cit+ variant after around 31,000 generations. Whether any new FI was actually generated is debatable, but if we grant that the two or three mutations actually increased FI we have a trivial level of FI generated, at best. Interestingly, the overall genome of the E. coli population has decreased by at least 8 percent, representing a very large genome-wide loss of FI for a very trivial gain of FI on a local level. Presumably, there are genes that are simply not needed within the experimental medium and these are being shed through a net deletional bias. More telling, over the current 50,000 generations, no sign of a novel protein family has begun to emerge which, when we look at the supposed evolutionary history of, say, hominids, and how many de novo genes are supposed to have evolved per 50,000 generations, spells serious trouble for neo-Darwinian predictions as to the rate of generation of de novo genes.
Re. ‘random’: in its strict sense, no such thing exists I suspect. In a strictly materialistic universe, everything is determined from initial conditions, even the evolution of wave functions in quantum mechanics. In science, when we say ‘random’ we are really referring to a situation where there are so many variables that we cannot predict the outcome. A crystal structure, say, NaCl is not ‘random’ but highly ordered. This is because the outcome is highly constrained by the laws of physics. The precise shape of a well-worn stone is not dictated by the laws of physics. The variables are so many that we can call the shape ‘random’ even though if we knew in advance all the millions of equations and variables that stone would encounter, we could precisely predict the outcome of its shape.
#104 Alan Fox: Regarding how rare functional biological proteins are, it is not the case that the only reliable way we have to estimate it is in the wet lab. Nature has been doing it since the OOL and we have a record of its experiments in the Pfam database available online. For example, I am working with Actin right now. The Pfam database has over 15,000 sequences but when I remove the redundant ones, I still get over 9,000 unique sequences for Actin. These sequences represent functional variants. I am assuming that non-functional sequences have been weeded out by natural selection. So we can regard nature as a gigantic laboratory where different sequences are being constantly tried out, eliminated or preserved according to whether they produce stable folding functional proteins. We can use the results to estimate a probability distribution for each amino acid at each position in the sequence (aligned array) and calculate the FSC according to the method published in one of my papers I cited earlier. Once we have the FSC, we can use Hazen’s equation to solve for the nf/nt ratio and nf itself to see how many different sequences might be functional in total. All the results indicate that functional biological proteins are extremely rare. Indeed, finding one is a needle in the haystack problem.
In this discussion, however, there is a difference between functional proteins and functional biological proteins. It is easy to randomly string together amino acids and get them to bind to something, as recent papers show. There is a tendency to call these ‘functional’ proteins because they bind to something. In biological life, however, it is not good enough to merely bind to something. Biological proteins have to bind to the right molecule, under the right circumstances, with the right strength such that the bonds can be broken under the right circumstances, etc. This requires a substantial fine-tuning of stable, folding functional biological proteins. This is why we see an enormous difference in the nf/nt ratios between so-called ‘functional’ proteins generated in the lab and actual biological proteins.
#105 Elizabeth B Liddle: I agree with you that the original Abel and Trevors paper assumed that there was no continuum as far as FSC is concerned. They also believed it could not be measured. I contacted them when I read their paper and suggested that FSC could be measured and that FSC could range anywhere from zero to very large values. The outcome of our discussion was our paper I cited in my original post that provides a way to measure FSC.
I also agree that it would be an irrational stretch to conclude that their paper establishes for all time that FSC cannot be produced by chance and necessity. Rather, I would say that they present a good argument and that our subsequent paper is the first step in testing their argument. It is only when we measure FSC for various genes, laptop computers, F-22 fighter jets, textbooks, and so forth that we are in a position to see if statistically significant levels of FSC can be generated by natural processes. That is were we must estimate the probabilities of various scenarios and go with the scenario that has a much greater probability than the others. When we look at the probability of an evolutionary search (natural selection/elimination coupled with mutation, deletion, insertions, genetic drift, crossover, etc.) producing the FI that codes for a stable, folding functional biological protein vs. an intelligent human producing the same level of FI, we observe that the a posteriori probability humans can produce that level of FI is 1; we do it every day. On the other hand, the probability that natural processes can do it can be carefully worked out from Pfam data and found to be typically less than 10^-200. A rational scientist will conclude that intelligent agency is a far superior explanation for the generation of statistically significant FI.
Re ‘choice’: In this discussion I am only interested in the standard philosophical definition of free agency … a choice is free if and only if a) it was not determined by any antecedent conditions and b) the agent could have chosen otherwise. You see a clear dichotomy between determinism and free agency. If a ‘choice’ is determined by the laws of nature, the initial conditions of the universe, and the entire resulting set of physic-chemical inputs of the moment, then it is not a free choice by the standard definition provided. It is not a ‘choice’ at all, but merely a deterministic outcome. True, the agent may believe it was a free choice that satisfies the two criteria spelled out above, but that belief is mistaken. She is only able to entertain it due to her inability to comprehend the sheer complexity of the myriad causal chains extending back to the moment of the birth of the universe. With this in mind, natural selection cannot perform ‘choice’. The outcome of any deterministic process of nature is determined, albeit by thousands or millions of variables, each of which was, itself, completely determined by the initial conditions, etc.
Re. taxonomy: Since life is determined in part by the digital FI encoded in the various genomes, and is quantized in packets such as promoter regions, genes, and so forth, it should be possible to classify life forms not by morphology but by similarity of the output of the FI system each one contains. We are not in a position to do this for small differences, but the discovery of orphan genes, and their uniqueness to certain taxa, gives us a more objective way to classify life. They effectively falsify the macro-evolutionary prediction that life has evolved in a continuum or series of very small steps. Instead, they verify the ID prediction that life was designed in ‘kinds’ (currently undefined, but science has the capability to delineate these kinds by focusing on the orphan genes as one key factor). I was once part of a team of engineers coordinating the build and test of experimental aircraft engines for Pratt & Whitney. A particular model would be designed, which then permitted a number of variants. The FI changes to produce the variants were small. If a substantially different aircraft engine was required, a totally new design was produced representing a quantum change in the FI needed to build it. As far as possible, similar parts were used, but new parts were also required. This is what we are discovering with life. Each design is capable of many variants through micro-evolution, however there are observed limits to micro-evolution. If we want something different enough, the FI is simply not there to produce it. A whole package of novel protein families may be required. They are designed straight up and downloaded into a novel genome exactly as we may be capable of doing someday in a lab in designing novel life forms. The tell-tale signature of a novel design is the the package of orphan genes along with their digital regulatory controls required to produce the new ‘model’.
Well, indeed. That is exactly what evolutionary biologists would point out. The evolutionary process of reiterated variation and selection does stumble upon new adaptations to changing niches constantly. And previous but less well-adapted solutions will, as you say, get weeded out. Bear in mind that these would not be non-functional, just not as good (in the context of the particular niche) as the solutions that out-breed them.
Well, no you can’t. Because you still have no idea what functionality lies in the unexplored and unknown vastness of as yet untried protein sequences.
#113 Alan Fox: The new adaptations you speak of within this context are new alleles within an existing protein family. In the Actin example I mentioned above, we have over 9,000 different sequences that all produce a functional Actin molecule with Actin’s unique 3D structure. This is microevolution and microevolution is entirely capable of producing novel sequences within existing protein families precisely because no change in FI is required. It is trivially easy. The OOPF problem I am talking about has nothing to do with variation within existing protein families and everything to do with the production of novel protein families.
Contrary to your assertion, we can indeed calculate the probability distributions for each amino acid at each aligned site in a protein family and estimate the amount of FI/FSC required to produce it. Those distributions begin to stabilize after 1,000 unique sequences, indicating that nature has done a pretty good job of sampling sequence space for a given protein family. It has nothing to do with how many other protein folds may exist in sequence space. If you will familiarize yourself with the method I published, or with Hazen’s equation, you will see that our ignorance of what lies elsewhere in sequence space is completely irrelevant to measuring the FI or FSC of a particular protein family. It does not even enter into the equation. When we do measure FI or FSC, we (i.e., science) find that the nf/nt ratio is extremely small, approaching zero. Indeed, in all other branches of science except for evolutionary biology ratios that small are regarded as zero. It is not just a few protein families that have infinitesimally small nf/nt ratios. Every one we look at is impossibly improbable.
One possible answer as to how an evolutionary search could have found them is that sequence space is filled with stable, functional folding islands as I think you are suggesting. Science, however, is consistently falsifying that prediction. The primary reason seems to be that biological functions are contingent on stable, repeatable 3D structures for at least 70% of all protein families, and perhaps even for the bulk of intrinsically unstructured or disordered proteins if we consider post-binding structure. Even though we could not possible test all of sequence space to see how common stably folding proteins are, all the work on protein folds indicates that they are extremely rare …. so rare that no one searches for them within a library of random amino acid sequences any longer. So how rare are they? Well, there has been work done on the total estimated number of protein folds and there are papers dealing with the taxonomy of protein folds. One estimate is that there are as many as 30,000 different possible folds. Let us assume that the average globular protein consists of three structural domains, each domain of which represents a particular fold configuration. This suggests that the total number of possible protein families could be as great as roughly 10^12. This is many orders of magnitude more protein families than are currently known in biological life. When we use Hazen’s equation to solve for nf, we estimate that for the average 300 residue protein family, there may be as many as 10^62 sequences for each family that are functional. (keep in mind this is an extremely generous estimate based on the assumption of site independence, which we know is not the case. Actual protein sequences have a very high rate of dependencies between sites, but let us be generous.)
From this, we can come up with a very rough estimate for the total number of stable, folding 3D sequences in 300 residue sequence space … roughly 10^74 sequences that will give stable 3D folds (this is very rough, but it will illustrate my point and help one see why scientists don’t search for novel stable 3D folds from a library of random sequences). One might think that 10^75 sequences is an enormous number, however, it is miniscule in comparison with 20^300, which is the total number of sequences in 300 –residue sequence space. This is why the theory that an evolutionary search, even if it involved all the planets in all the galaxies of the known universe, is utterly implausible. Keep in mind that it has also been shown that natural selection is inoperative in non-folding sequence space since the sequences in that region are non-folding and non-functional. In fact, natural selection tends to eliminate misfolded or non-folding proteins as they tend to form amyloids which can greatly reduce the fitness of an organism. If you wish to embed the coding for folding functional protein families into the genomes of life, you will need something much more powerful that an evolutionary search that plods along at physico-chemical speeds. You will require intelligence for any statistically significant level of FI to code for biological proteins.
Current experiments, as I mentioned earlier, that find random sequences of amino acids that bind to something are toy examples of functionality for reasons I mentioned in my previous post. They are interesting, but what we are really concerned with is biological life. When we measure the FI required to code for the average biological protein family, it indicates a high level of intelligence and design was required to carry out their functions. Genetics is beginning to make that increasingly obvious when we acquaint ourselves with the molecular computers and molecular machines operating within the cell, each of which must be coded for within the genome so that the right genes are activated in the right sequence and the right components are transported to the right location for assembly to occur. The belief that biological life came about through mindless natural processes is beginning to look like the most bizarre creation myth humanity has ever fallen for, if I may be so bold as to say so.
To all: it is near quitting time on Friday, with the long weekend coming up here in Canada. I make a point of having a life on the weekends, so I will not be reading or responding to any further posts until next Tuesday, when I get back to work. Sorry for the delay, but it is time to howl ….. woo hoo!!!!
KD, thanks for your OP and comments, they’re truly appreciated. Enjoy your long weekend. 🙂
How can Chance appreciate anything?
=P
Alan Fox:
I’ll understand if you’re not capable of keeping up, but that was not the claim Elizabeth made.
For Alan:
Mung:
Elizabeth:
Mung:
Elizabeth:
Elizabeth Liddle:
truer words …
Mung @117, I stumble upon appreciation about 1 in 10^74 occurrences or so. :D
Thank you for the lack of information provided.
In this thread, I have enough on my plate absorbing Dr Durston’s comment 114 (Thanks for a full and considerate reply to Dr Durston). If you want to make some point that you think may be of interest, I suggest you tell me what issues are, rather than what they aren’t.
Dr Durston @ 114
I’d like to focus on the functionality of unknown proteins (neither yet synthesized nor currently found in nature – noting this causes me to reflect on proteins that existed in extinct organisms, mostly a closed book to us – except where DNA traces remain) Just briefly, having read what you write several times, I still don’t see your justification for saying anything about this enormous group.
I would agree that, were the whole of protein sequence space very sparsely dotted with potentially functional sequence”s in a sea of non-functionality, darwinian mechanisms might be found wanting. But we cannot make that claim because we can say nothing about hypothetical sequences. We need nature or the lab to test examples first. When you can reliably predict the properties of a protein before synthesis such that synthesizing confirms predicted properties, maybe then we can conclude whether unguided evolution can find solutions by stumbling upon them.
#123 Alan Fox: I think the hypothesis you are suggesting is that functional sequences may be common enough in sequence space such that an evolutionary process would be reasonably likely to find a sufficient number of different functional proteins to enable the full diversity of life we see today.
This is a reasonable hypothesis. Anyone interested in Darwinian theory or ID theory needs to seriously consider it. Furthermore, recent work suggests that randomly sequenced non-biological proteins can bind to particular molecules, thus qualifying as ‘functional’ so far as their ability to bind to something. Now let us consider that hypothesis a little more carefully.
First, the concept of ‘function’ is highly system-dependent. We can decree in a lab that if a random sequence of amino acids binds to something, then it is ‘functional’ at least so far as fulfilling the function of binding to something. Let us suppose that roughly 9 out of 10 random sequences can bind to something if we experiment long enough. If merely binding to something was all that biological life required, then nf/nt is very close to 1, and using Hazen’s equation, the amount of Functional Information (FI) required to obtain these sort of functional proteins would be very close to zero. This is trivial and no intelligence would be necessary to locate them.
There is a problem, however. It appears that many proteins without a stable folded 3D structure tend to clump within the cell, forming amyloids. These are usually detrimental to the cell and can eventually be lethal to the organism. Therefore, we need proteins that are both functional, but do not clump. One advantage of globular proteins is that their stable 3D structure helps safeguard against clumping together in to amyloids.
There is another problem. There are many compound protein structures that appear to be assembled from a number of different proteins, each with their own unique 3D structure, fitted together to form molecular machines. F1 ATPase is an example of a rotary motor with different components that not only nicely fit together, but which also have the right physico-chemical properties to rotate. The famous flagellum is another example. Another is the amazing Ribosome. So mere binding is as insufficient as randomly gluing rocks and sticks together to get some sort of super sonic fighter jet.
All this to say that the requirements of biological life might require a little more sophistication to its protein families than merely binding to something. If we just consider the problem of clumping, I think it eliminates most of sequence space … but I don’t know this for sure. We do know two things, however …
1. about 70% of protein families found in biological life appear to be globular in structure (i.e., have a stable, repeatable folded 3D structure)
2. stable 3D structured proteins are extremely rare in sequence space … so rare that scientists recognize it as futile to randomly search for them. My own work coupled with work done on estimating the number of different protein folds, indicates that these stable, 3D structured protein families are so rare that a mere 10^43 searches (an extreme upper limit for the total number of mutations/insertions/deletions that have occurred in the entire history of life) would be at least 200 orders of magnitude too few to hope to discover even one average sized (300aa) 3D-structure protein family in 4 billion years of evolution.
Yet life seems to have ‘discovered’ tens of thousands of them. Something is going on. If you see the baboon down the street winning the lottery thousands of times, you know that somewhere in the processes, an intelligent person is fudging the system. Of course, a super intellect capable of running the calculations for all the different amino acid sequences, torsion angles, folding dynamics, and so forth would not have to randomly sample sequence space. All the different desired shapes and interactions for the overall design could be input, the equations run, and the proper sequences identified. To me, this is the overwhelming rational option. No sane person who, having understood the design requirements for even the ribosome, could for an instant even entertain the absurd notion that some sort of evolutionary search could cobble that together, forget about an entire cell with all its machinery. Natural selection (elimination) is an extremely crude ‘guide’. As my Darwinist prof said in our graduate Genetics Algorithm course, ‘when we began to design algorithms that modeled real world evolution, they never did anything.’ The fitness function is far too crude to guide a search into ‘finding’ stable 3D structures … forget about building molecular computers and molecular machines and integrating everything into a fully operational cell that predicts its own fate according to the inputs into its own computing system.
Another way of looking at your hypothesis: If 9 out of 10 random sequences of amino acids perform some sort of binding function, then why is it that life seems to have ignored them and gone for a set of tens of thousands of globular proteins which we know (and all our science verifies this) are mind-boggling rare in sequence space. We would predict that if 90 percent of all random sequences are functional (or some other decent percentage), then those would be the sequences that would consistently show up in an evolutionary search, not sequences that have a probability of 10^-300. How is it that biological life ‘found’ thousands of extremely difficult sequences in only a maximum of 10^43 trials? Any way we look at it, the search for stable 3D structures is a needle in the haystack search for which evolutionary algorithms are useless. There is no fitness gradient for selection to be operative.
Summary Statement: In virtue of our observations that life appears to require protein families that have a stable 3D structure, and given our knowledge that 3D structures that are stable are extremely rare in sequence space, so rare we could never hope to find one even with today’s computational resources, we must conclude that the kind of proteins life requires are far more sophisticated and rare than mere random sequences of amino acids that bind to something in the lab.
We have it on good authority that the first system capable of Darwinian evolution was really quite simple. No one know what such a system would be like, and there is no underlying theory, but it must have been so.
So your objection fails.
And once we have even the simplest system capable of Darwinian evolution, all else follows.
That’s the consensus of all modern science, religious fundamentalists and ID “theorists” notwithstanding.
Hi Kirk, I’ll try to get semi-serious recognizing that you are an actual practitioner in the field.
Is it possible that modular protein domains are the answer to the dilemma (assuming there is a real dilemma)?
Simple domains were discovered by random search and then these successful protein “parts” were then re-purposed into the wide variety of proteins we have today.
Do you know of anyone who believes this hypothesis and in your opinion why this an untenable hypothesis?
Thank you for visiting UD!
#126 Mung: The only reason I can see that the modular theory is entertained is that few, if any, have actually crunched the numbers to see if it makes the discovery of globular proteins more probable. There are two problems with this theory. First, the smallest ‘step’ or module that might be able to stand alone is a single structural domain, typically consisting of (very roughly) 100 amino acids plus or minus. This is a massive step and the probabilities are so small that we cannot expect the universe to have stumbled on more than one or two single structural domain modules over the past 13 billion years. In my first paper, I published the results for several single-domain protein families. If you take the FSC value and solve for Hazen’s nf/nt, you will see that the modular theory is a no-go.
The second problem is that in the modular theory, we can transform the OOPF problem for the average globular protein from an average of 300 ‘letters’ each of which has 20 very probable options (assuming we have a mechanism to produce biological amino acids) to an average of 3 ‘letters’ each of which is extremely improbable (where the average globular protein has three structural domains). If you approach this problem using the equation for FSC published in my first paper (cited in my original post), you will see that the probability remains the same whether we use the highly probable amino acid ‘letters’ or transform the problem into the highly improbable structural domain module ‘letters’. There is no way to get around the probability problem unless the Problem Solver has sufficient knowledge of the laws of physics and, coupled with the desired design requirements, knows how to construct structural domains. In that case, probabilities are no longer a consideration; they are replaced with units of functional information. An intelligent mind can replace probability problems with FI solutions.
Yes. Except I don’t think of the process of evolution as a search for anything, rather stumbling upon variations close to the primitive protein. Evolutionary processes don’t take huge leaps in the dark.
I agree. Jack Szostak’s work with ATP binding showed the way.
OK
Your first point is context. And I take the point. But remember that nothing rules out co-evolution. The Lenski experiment (still ongoing) showed that is possible with the citrate digestion ability that needed two successive mutations.
But we are back to the same point. You will have to clarify how one can make a claim about unsearched sequence space when it remains, well, unsearched. I would be amazed if blind search of sequence space will be any more productive than monkeys on typewriters. The holy grail of biochemistry would be a way of predicting functionality without having to synthesize and examine the properties of a previously unknown protein.
I’d be most interested in the evidence and methods that brought you to that conclusion. Is it in a specific paper that is available on-line?
I can’t rule out the possibility that I am insane and nobody around me has noticed but there may be others who have still not given up on the possibility of evolutionary explanations. Three billion years and unimaginably large populations of rapidly turning over prokaryotes is a big laboratory in which to test the basic chemistry, which all seems to have been in place proir to the diversification into sexually reproducing eukaryotes and beyond into multi-cellularity.
We don’t know what proteins were in use by organisms that have left no trace of DNA to examine. For all we know, different proteins were in use and were superseded. Regarding globulins, a paper that covers the issue of how to estimate their rarity as functional proteins would be most appreciated, as noted above.
I don’t think we can be so cut-and-dried. For one thing, we don’t know what transitionals may have existed back in that 3 billion years or so for which we have no direct evidence.
But thanks for taking the time to respond and I’ll look out for any further comments.
Showed the way to what?
It already had that ability- to digest citrate. The mutations just allowed it to digest citrate in an aerobic environment. And Lenski still hasn’t shown that the mutations were random in any sense of the word.
What Lenski has demonstrated is evolutionary processes are very limited. No new proteins, no new protein configurations, nothing that can be extrapolated into macroevolution.
KD,
Thank you very much for taking the time to respond to my questions.
Is your first paper reproduced in Able’s book The First Gene? If not, and I ask your indulgence if you’ve already mentioned it earlier in the thread, to which paper are you referring?
I’m moving, my book on modular protein domains is in storage, if you’re still around when I get my internet service back and can access that book again perhaps I’ll have a followup.
Until then, as a biophysicist, have you had a chance yet to read Life’s Ratchet?
If so, care to comment?
BTW Dr Durston, I’d be interested to hear if any cross-pollination has occurred between you and Janet Thornton and her group at Cambridge. This paper looks interesting.
Mr Fox, I’m fairly certain that you have already been corrected before on Szostak and Lenski’s empirical research and that you now simply refuse to accept falsification from the experimental evidence, but just to make 100% certain that you, and others, know for a fact that you are making completely bogus claims in regards to the evidence at hand, I will remind you:
This following paper was the paper that put the final nail in the coffin for Szostak’s work for ATP binding:
Here is a very interesting comment by Nobel prize winner Jack Szostak himself:
Mr. Fox will you do the proper thing by now acknowledging your mistake and gratefully thanking those who corrected you on your error? Or will you, as usual, ignore the correction only to wait and make the same bogus claim down the road when you think no one who knows your error is looking?
I don’t believe Gregory and his like-minded believers in thaumaturgy by inanimate matter realise that empirical science is no more and no less than a natural (no pun intended), indeed, inevitable, development from the utterly unique corpus of knowledge and understanding of the Christian Church accrued over more than 1000 years (now 2000).
Indeed, add to that, its epistemological matrix, Judaism, equally unique in OT times, and it is clear that what we have been witnessing – despite the farce of the so-called, Enlightenment – is theism first, atheism nowhere; which is why those students interviewed by Ray Comfort were flummoxed to think of a single famous atheist.
The systematic study of the world, as has been pointed out many times, is one of the benefits granted to mankind by Christ’s Incarnation, through his church’ So, no, the rise of science did not have to await the arrival of atheists.
On the contrary, it was in Elizabethan times that the Protestant Christian, Francis Bacon proposed a new science of observation and experiment to replace the traditional Aristotelian science, as Descartes, a Huguenot, was to do later.
The atheists have always lived off the back of the real scientists, theists and, of course, believers in Intelligent Design. Capitalizing those two words is deeply offensive to me, since it is no more than the most elementary common sense that the worked was designed – and that, by a super intelligence, beyond our imagination.
#130 Mung: The first paper I referred to that presents a method to measure the Functional Sequence Complexity (FSC) of protein families is here . I’ve not read Life’s Ratchet but just reading the brief summary, he seems to be starting with some simple molecular machines which ‘custom-build’ other molecular machines, etc. When I read stuff like this, it is discouraging as to how low science has stooped in confusing creative story telling with doing actual work. I have read tons of stuff over the past 25 years that boils down to blissful, naive optimism which refuses to face the cold hard reality of how difficult it is to build nano-machines by molecules ‘crashing into each other’. The gigantic Achilles heel in all this just-so story telling is a failure to actually crunch the numbers. The very first step one ought to take before waxing eloquent with the latest evolutionary scenario is to test the hypothesis that the particular protein structure being extolled can actually be stumbled upon in a blind evolutionary search. More on this below by way of example.
Alan Fox: I agree with you that the process of evolution is not a search for anything. The term ‘search’ is a term we use in Genetic Algoritms (GE’s), which is our best computational tool to attempt to model real-world evolution. But you are right in that the real world evolutionary process is a blind, mindless process.
I also agree with your statement that ‘evolutionary processes don’t take huge leaps in the dark’, but I think you have essentially falsified the neo-darwinian theory of macro evolution when you apply that premise to the gigantic leaps it would need to take across non-folding sequence space to ‘locate’ (by blind, mindless random walks) biological protein structural domains (if one wishes to take a modular approach). Blanco (Blanco FJ, Angrand I, Serrano L (1999) Exploring the conformational properties of the sequence space between two proteins with different folds: an experimental study. J Mol Biol 285:741-753) showed this in his attempt to evolve one protein into another.
Lenski’s work in the LTEE is highly commendable. Finally, after 150 years of granting the conclusion before the experiment has been carried out, we are actually witnessing the capabilities, or lack thereof, of blind, mindless evolution at work. Two successive mutations are almost certain going to occur. That is trivial. For the average structural domain, we are going to need approximately 100 successive mutations before mindless evolution ‘lucks’ into a stable, repeatable structural domain. No structural protein biologist thinks that stable 3D structures are common in protein sequence space. The consensus is that they are extremely rare. My objective was to figure out how rare they are. The first step is to measure the amount of FI or FSC required to code for any member at all of a protein family. The real story in the LTEE is that it is falsifying the neo-darwinian hypothesis that novel protein families were produced at a rate of several per 50,000 generations. The LTEE has serious negative implications for the evolution of larger organisms that have a much slower replication rate. This, however, is a major discussion in itself, worthy of a future post on this forum.
An example of a single structural domain target size (and the kind of evolutionary leap it would require) is given by Ribsomal S12. I published the FSC required for Ribosomal S12, a single domain protein that is absolutely essential for life here . It is a universal protein found in all life forms, and is conjectured, therefore, to be a component of LUCA. The FSC is 359 functional bits. By plugging that into the FI variable of Hazen’s equation, you can solve for nf/nt. It turns out to be about one chance in 10^-108. Evolution does not have an ‘unimaginably’ large number of chances. A published upper limit is 10^-43. So evolution was fortunate enough to mindlessly find a target absolutely essential in its shape/structure to enable the greater function of the Ribosome, and it mindlessly performed that fantastic leap on a ‘target’ with frequency of occurrence in sequence space of about 10^-108 in less than an absurdly small 10^43 trials. Darwkins’ climing Mount Improbable is a myth when it comes to stable folding domains. Folding is such that it is all or nothing. My more recent work here shows just how integrated those structures are.
If you are aware of any paper at all that shows that stable folding proteins are relatively common in sequence space (by ‘relatively common’ let us use 10^43 searches, which is an upper limit), then let me know about it and we will discuss it. The consensus is that they are so rare, it is useless to search for them. Thus, we have turned to reverse engineering biological proteins, figuring out their design principles, and intelligently designing them using these principles …. Intelligent design in action in the lab. See, for example Fisher MA, McKinley KL, Bradley LH, Viola SR, Hecht MH (2011) De novo designed proteins from a library of artificial sequences function in Escherichia coli and enable cell growth. PLoS ONE 6:e15364 and count the number of times ‘designed’ is used (29 times), ‘design’ is used (11 times) and ‘strategy’ is used (3 times).
With regard to ‘transitional’ protein families that may have occurred in the past, all our science is showing is that there are no such thing as ‘transitional’ structural domains. Structural domains occupy what we can refer to as clearly bounded islands in sequence space. Each domain can be regarded as a ‘fold-set’ which can actually be capable of 2 or 3 unique, distinct folds depending upon meta-stable states in folding, (itself a design masterpiece), but each island is surrounded by unstable folding space, shown in many papers as early as Fisher MA, McKinley KL, Bradley LH, Viola SR, Hecht MH (2011) De novo designed proteins from a library of artificial sequences function in Escherichia coli and enable cell growth. PLoS ONE 6:e15364 and more recently in Yang JR, Zhuang SM, Zhang J (2010) Impact of translational error-induced and error-free misfolding on the rate of protein evolution. Mol Syst Biol 6:421
Summary Point: At the end of the day, real science has to go with what is testable, verifiable and falsifiable. Appealing to what we might discover, or what we don’t know to justify belief in neo-darwinian macroevolutionary theory, especially in the face of real science which is making such beliefs increasingly untenable, is not science. Real science indicates that the average stable structural domain families are extremely rare in sequence space requiring gigantic leaps in any supposed mindless evolutionary process, essentially falsifying the theory. A hyper-intelligence, on the other hand, is capable of understanding physico-chemical properties, solving millions of equations simultaneously, containing hundreds of thousands of variables (including the torsion angles), and with the results, creating the right-shaped structures to fit together into molecular machines such as the ribosome or F1ATPase and building an entire organism. We do similar all the time with much simpler structure such as laptop computers, and spacecraft. It is all ID in action.
I have looked at Thornton’s work, but her group is focused on the evolution of functions of existing protein families and super families. My work focuses on the structure of protein families. Each structure or each protein family may have a variety of functions and two or three fold-sets. Indeed, a good design will choose those structures that are multi-functional, or within a short evolutionary step of novel functions. From what I gather, her work explores the evolutionary capabilities of existing structures/families. Those are relatively easy for an evolutionary process to stumble upon. Stable structural domains are not.
AF:
IANAS, but isn’t there something like a principle of uniformitarianism where, absent evidence to the contrary, one can infer that what is true about what has been sampled or searched will be more likely to hold true for what has not yet been sampled or searched than the opposite?
I would think it would be quite routine to make claims about what is unsearched based on what has previously been discovered. But again, IANAS.
@ KD
Thanks for the excellent guest post!
#136 Optimus: It is an honour and a pleasure.
KD,
Yes, indeed, thanks for posting. Most interesting.
Please come back often.
I second that, please Mr Durston, please post as often as you can. Thanks.
@ Upright Biped
Dr. Durston has been awarded his Ph D in Biophysics at Gueph, now, I believe.
@ Dr Durston
Thanks for bringing the work of Michael Hecht and his team to my attention. His paper does not seem to support your claims. Quite the contrary.
Why is it quite the contrary? Or does not support Dr. Durston’s claims?
#141 Alan Fox: You will need to support your assertion with some details. Which of my claims does the Cherny et al. not support and what, specifically, in the paper are you referring to? I trust you are not referring to the designed (i.e., ID in action) alpha-helix proteins they used in their experiment. The 4-helix bundles were designed using information we have reverse engineered (design derivation, another way ID is firmly entrenched in science). They were not, however, designed to bind to anything in particular. There is nothing in this paper that supports your claim. If you believe there is something I have missed, please do point it out. Did you not notice the word ‘designed’ they use to describe how they achieved the 4-helix structures? Please note that I have never contested the ability of random sequences to bind to something.
Dr Durston at 143
My point is simply that you can’t infer that unknown sequence space is rare in functional proteins. One just doesn’ t know. Maybe evidence will emerge but, for now, we just don’t know.
Alan, Dr Saurer did the experiments and demonstrated sequence space is rare.
Again your ignorance, while amusing, means nothing here.
#144 Alan Fox:It appears to me that you are confusing function with stable 3D structure … ‘function’ is common if any sort of binding qualifies; ‘structure’ is extremely rare and all our science verifies that. If you will review my posts, I have consistently argued that functional sequence space for proteins is probably extremely common, provided we are willing take a loose definition of ‘function’ to include binding to various molecules and not worry about stable 3D structures. An amyloid of misfolded proteins sticking to the side of my coffee cup would qualify as ‘functional’ under this looser definition, but that is irrelevant to how stable, functional 3D folds were ‘found’. What I have also argued for is that proteins with a stable 3D structure are extremely rare in sequence space, and that is what 70% of biological proteins seem to be. The Cherny et al. paper does nothing (e.g., absolutely nothing) to support the idea that stable 3D folds may be common in untested sequence space. In fact, it does quite the opposite; they relied on designed 4-helix barrel structure. This is explicitly stated in their paper. There are other papers that expand on how we designed this structure using knowledge we have reverse engineered from biological proteins.
“Structure” is extremely rare in proteins? What on Earth do you mean by that?
AVS,
What he means is, that for new structures to emerge in evolutionary history new structural proteins are required. That should be fairly non-controversial.
These structural proteins are extremely rare in sequence space. period. fact. and all our science verifies that.
New structural proteins require new structural folds. So imagine all the “functionality” you want to conjure up, it’s all irrelevant.
What’s relevant is protein folds.
But it’ll take you longer than your life span to stumble on one through random trials. IOW, Darwinism is falsified by the facts.
cheers
This is wrong. The rarity of functional proteins in the totality of possible protein sequences is unknown.
From a Wikipedia article De novo protein structure prediction;
And that was just for one protein sequence.
As I keep saying, the claim that functional proteins are rare is, at best, premature. This, of course has major impact on Dembski’s CSI concept and P(T|H). As Dr. Liddle points out “P(T|H) is the Probability that we would observe the Target (i.e. a member of the Specified subset of patterns) given the null Hypothesis”. I’ll grant you, demonstrate that functional proteins are needles in haystacks and that will be a real problem for evolutionary theory.
#149 Alan Fox: Again, you are confusing ‘functional’ with ‘structural’. In your post, re-read the opening quote you used from me. You will see that it says ‘structural’ not ‘functional’. Biological life seems to require STRUCTURE for its functional proteins. (not sure of your background, but if the importance of structure in biological protein function is something you are unfamiliar with, you may want to read a couple of quick articles on the web on this subject.) So in this discussion, we not merely interested in some random ‘function’ that a scientist in the lab defines, such as ‘will it bond to something’? In this discussion, we are interested in proteins that will fold into a stable, repeatable 3D structure as our first step. Then, we can sift through those to see which ones might be functional.
Second, your Rosetta quote makes my point. The Rosetta results were for an existing sequence that we already knew produced a stable 3D structure, we just had to solve the structure. We gave it the sequence (T0283) and then we tried to figure out its 3D structure. An entirely different problem ….. extremely difficult, but vastly easier than feeding in random sequences to see if they have any 3D structure.
Third, Dembski’s P(T/H) is not what we are discussing here. We are not discussing the probability of observing some particular pattern/structure/sequence. The probability that Hazen et al use in the paper cited eaerlier on in this discussion is the ratio of ALL the functional sequences for a given biological protein, observed and not-yet-discovered (please notice the ‘not-yet-discovered’ bit), over the total number of possible sequences for the given number of sites in the sequence.
Finally, nobody in the field thinks that stable, 3D structures are common in sequence space. Please note I am talking about 3D structural proteins, which biological life seems to need, not some random sequence that has a lab-defined ‘function’ of merely binding to something. I repeat, nobody in the field thinks that sequences forming stable, 3D structures are common in sequence space. If you believe they are, show us the paper. In science, we need to operate on the basis of what is testable and verifiable. The hypothesis that 3D structures are common in sequence space is consistently falsified in experiments done in the field. To clarify, we have tested it numerous times and consistently failed to verify it. Everything we are learning about protein sequence space is falsifying the hypothesis that stable, repeatable 3D structures are common. This, in turn, is falsifying the neo-darwinian hypothesis.
All: I am taking two weeks vacation starting today, so I will not be able to respond to any further posts until early September. With this in mind, I think I will leave the last word to Alan Fox and others who wish to respond.
KD:
Well noted.
I think it is an inadvertent testimony to the force of the needle in the haystack point that objectors are being reduced to imagining that folding AA sequences yielding key-lock fitting patterns in life forms are common in sequence space.
This seems to be one of those, this must have happened for what I “know” to be true so it is also true.
What is missing of course as you point out is proper empirical warrant.
KF
Enjou your break, Dr. Durston.
Oops
Enjoy!
I, for one, hope to see more of Dr. Durston here. This post (and its comments) was awesome.
Anyone want to help craft a really interesting question for Kirk to ponder upon his (hopeful) return?
Dr. Durston,
This thread has disappeared down the list of posts on UD but I have a question/request for you based on some recent comments on other threads.
It has to do with ORFan genes. Supposedly, they are quite common in each species’ genome. However, from the conversation that takes place when they are discussed it is not clear just what they are of how much is known. Assume my knowledge is limited for the questions that follow. Some questions,
Are ORFans really genes? I assume a gene is something that actually codes for a protein.
Can something be called a gene if it is just transcribed but never translated into a protein?
Are ORFans just genomic entities that have mutated over time and may not actually do anything functional in the genome? I assume that the term “junk DNA” is becoming less appropriate but that there are some parts of the genome that are not functional in any way.
Are there ORFans that are in some ethnic groups that are not in others. Let me give some examples.
I recently had my genome analyzed by 23andme and it indicated that about 3% of it was due to a Neanderthal background. From what I understand this is common as the average for non-Africans is 2.7% Because of this I would expect that these differences are just different variations of alleles but are they more than that.
Would there be different ORFans in Australian aborigines than in African? The Australian aborigines supposedly migrated there 40,000 years ago and so would be one of the most isolated groups on the planet. Or do all humans have essentially the same genes and what we see in both ORFans and other genes is just different alleles?
Finally, could these ORFans be analyzed for the protein generating capability. I probably poorly understand proteins but is it possible to analyze an ORFan to see what kind of protein it would be and whether it would be functional or not.
If you should ever read this and have time, your answers would be quite interesting to many of us here.
jerry:
Are ORFans really genes?
Yes, they are actually genes.
Of course, ORFan is a play on words. ORF – Open Reading Frame, Orphan – without parents or of unknown parentage.
I posted a number of relevant posts (including reference links) which Salvador Cardoza then proceeded to delete in the interests of “Serving the Intelligent Design Community.”
Or maybe it was out of pure pettiness.
I’m going with the latter.