Two forthcoming peer-reviewed pro-ID articles in the math/eng literature

_{William Dembski

January 20, 2009

Informatics, Intelligent Design}

Share: Facebook; Twitter; LinkedIn; Flipboard; Print; Email

The publications page at EvoInfo.org has just been updated. Two forthcoming peer-reviewed articles that Robert Marks and I did are now up online (both should be published later this year).*

——————————————————-

“Conservation of Information in Search: Measuring the Cost of Success”
William A. Dembski and Robert J. Marks II

Abstract: Conservation of information theorems indicate that any search algorithm performs on average as well as random search without replacement unless it takes advantage of problem-specific information about the search target or the search-space structure. Combinatorics shows that even a moderately sized search requires problem-specific information to be successful. Three measures to characterize the information required for successful search are (1) endogenous information, which measures the difficulty of finding a target using random search; (2) exogenous information, which measures the difficulty that remains in finding a target once a search takes advantage of problem-specific information; and (3) active information, which, as the difference between endogenous and exogenous information, measures the contribution of problem-specific information for successfully finding a target. This paper develops a methodology based on these information measures to gauge the effectiveness with which problem-specific information facilitates successful search. It then applies this methodology to various search tools widely used in evolutionary search.

[ pdf draft ]

——————————————————-

“The Search for a Search: Measuring the Information Cost of Higher Level Search”
William A. Dembski and Robert J. Marks II

Abstract: Many searches are needle-in-the-haystack problems, looking for small targets in large spaces. In such cases, blind search can stand no hope of success. Success, instead, requires an assisted search. But whence the assistance required for a search to be successful? To pose the question this way suggests that successful searches do not emerge spontaneously but need themselves to be discovered via a search. The question then naturally arises whether such a higher-level “search for a search” is any easier than the original search. We prove two results: (1) The Horizontal No Free Lunch Theorem, which shows that average relative performance of searches never exceeds unassisted or blind searches. (2) The Vertical No Free Lunch Theorem, which shows that the difficulty of searching for a successful search increases exponentially compared to the difficulty of the original search.

[ pdf draft ]

—————

*For obvious reasons I’m not sharing the names of the publications until the articles are actually in print.

Comments

CJYman @151

Thanks for the detailed reply. It appears that I do understand you correctly. In order for Dr. Dembski to present these two papers as support for ID, though, he (or someone) needs to make the links you hint at explicit.
I see where you are coming from and I think that the reason he doesn’t make the connection explicit is because it seems that these papers are more of a response to inflated claims about how evolutionary algorithms can account for biodiversity. These papers were merely showing that the extent to which an evolving system increases probabilities is proportional to the improbability of finding that system (evolutionary algorithm) in the first place.
I still don't see this from the papers themselves. When I saw the title about peer-reviewed papers supporting ID, I was excited. After reading them, I can see it will be all too easy for ID opponents to dismiss them as having no biological relevance. I would love to see a third paper that ties this all together.
The papers [along with the NFLT] explicitly detail the fact that in order to increase probability of generating a given pattern, a set of laws matching search space structure and search procedure, which is even more improbable, must first be generated. Thus, evolution is not, as some have claimed, a free source of information — information being an increase of probability. This increase of probability, labeled active information in the paper, is merely pushed back to higher and higher levels.
Again, this argument seems to be that the physical laws of the universe are designed, but that, given them as they are, evolutionary mechanisms are sufficient to explain biological diversity. I suspect this is not Dr. Dembski's position. ;-)
What the paper does show is that this increase in probability seen in the operation of evolution still needs an explanation. Evolution is not an explanation of its own success at increasing probability anymore than an EA can create itself absent foresight.
That's great, but it still seems to cede the field of biology to modern evolutionary theory.
If a computer program based on only chance and law (absent previous foresight) won’t produce CSI, why would an infinite amount of such programs be a better explanation rather than foresight, which is indeed characterized by its generation of CSI?
An aside due to my inner geek: Is there a formal proof that CSI cannot be generated from evolution simulations?
IOW, all simulations of evolution require foresight so, barring this universe being the only option, on what grounds could someone say that the "real deal" requires no foresight as a scientific fact or that there is "no evidence for teleology within life" or that "ID is unscientific?"
I don't believe that "all simulations of evolution require foresight." We've discussed several different genetic algorithms in recent threads here and Dawkin's Weasel is the exception rather than the rule. I'm not sure how this is relevant to the two papers, though. JJJayM_{January 27, 2009
January
01
Jan
27
27
2009
11:41 AM
11
11
41
AM
PDT}

gpuccio @150

This overstates the case somewhat. While some full proteins are species specific, most do have underlying partial homologies. It is those that are used as evidence for common descent.
And so?
The underlying partial homologies are evidence that supports MET mechanisms of incremental change. You can't consider just the full protein as though it came into being all at once. Because even those species specific proteins have significant overlap with proteins in other species, evolutionary biologists can and do consider them consistent with MET.
Even if you group the proteins which have strong homologies, you still have myriads of separated islands.
Could you point to some literature cites that support this claim? I have looked for such information for quite some time, because it seems that this is a very promising area for ID research. I haven't found any clear results that show that genome space is fragmented in this way.
Let’s take the example of myoglobin. At some point in natural history, this protein appears. It is about 154 aminoacids long. Where did it come from?
An evolutionary biologist would probably say it came from a protein that was about 153 amino acids long. Or from a very slightly different protein that was about 154 amino acids long. Unless ID researchers can show this is a mathematical impossibility, the MET mechanisms remain scientifically credible.

I’m afraid that “extremely varied” isn’t supported by the empirical evidence.
As an answer, I quote again from my previous posts: from my post # 52: "The proteins we do know (and we know a lot of them) are really interspersed in the search space, in myriads of different and distant "islands" of functionality. You don’t have to take my word for that. It’s not an abstract and mathematical argument. We know protein sequences. Just look at them. Go, for example, to the SCOP site, and just look at the hyerarchical classification o protein structures: classes (7), folds (1086), superfamilies (1777), families (3464). Then, spend a little time, as I have done, taking a couple of random different proteins from two different classes, or even from the same superfamily, and go to the BLAST site and try to blast them one against the other, and see how much "similarity" you find: you will probably find none.
While I appreciate your effort, that isn't a particularly rigorous process for tracking homologies. You might identify a new homology with that approach, but failure to find one after a handful of searches doesn't invalidate the entire corpus of evolutionary molecular biology. Identifying proteins that are a result of a genome in an unreachable area of genome space will require extensive, painstaking work to identify candidates and trace them back to their progenitors in a recent common ancestor with another species. Taking humans as an example, if MET mechanisms are sufficient you should see similar molecular constructs, for a certain level of complexity, in chimpanzees. Other great apes should show less similarity, and other primates less still. If you find a relatively complex construct (protein, gene, etc.) without a corresponding similarity, you'll have a solid candidate for further research. I would love to see ID researchers taking this tack.
In the genome of E. coli, a well known bacterium, we can find at least 2018 proteins described, grouped in 525 groups of similar proteins. The biggest group includes 52 proteins, while the smallest groups include only two proteins each, and there are 291 of them. Do you still think that "extremely varied" is not supported by the empirical evidence? What empirical evidence are you suggesting?
There are certainly a lot of different proteins, but again you can't assert that they can't be arrived at through MET mechanisms without looking at the probably precursors and overlapping sequences in other organisms. MET does not claim that the whole protein came into being in one fell swoop. We have to address what MET actually says, not what we wish it said.

Again, we in the ID camp need to be very careful to address the real positions held by ID opponents
That’s exactly what I am trying to be.
rather than strawmen that not only can be easily knocked down but that can also be used to distract endlessly from our core message.
I don’t think I am using any strawmen. You can see that I am very precise and explicit in my arguments and in my answers, trying to address exactly what others, including you, are saying. If sometimes I misunderstand what the other is saying, I am ready to apologize.
Your work with BLAST does not, unfortunately, support your claims. Instead of looking at entire proteins or genes, you must consider subsequences shared with likely precursors and other descendents from those precursors. This is exactly what molecular biologists and bioinformaticists do on a daily basis. Numerous homologies have been identified and the literature contains more every day. I will be delighted if you have really identified the disconnected islands that disprove MET, and if you have you should definitely publish your results, but I don't see support for that in what you've presented so far. ID opponents make some strained arguments, but one that is valid is that anyone who discovered what you claim here would be world famous. It stretches credibility to think that everyone with access to the same tools you used could be Expelled and prevented from publishing. JJJayM_{January 27, 2009
January
01
Jan
27
27
2009
11:27 AM
11
11
27
AM
PDT}

Gpuccio This business of frameshifts is rather interesting. I wish I had time to really understand the papers. But there is a broader thing that concerns me. I can't see that frameshift is a Darwinian concept. It is just a way generating a DNA string from a parent that gives a radically different protein string. All other known methods, it appears, do not generate radically different proteins. So if we come across a protein which is radically different from others and there is no evidence that frameshift was involved, then there is a problem knowing how that protein was generated. But it is a problem for all theories that assume DNA is created by modification of parent DNA. Unless you believe God inserts complete DNA strings into cells from time to time then, this equally a problem for a non-Darwinian theory. In fact frameshifts could be interpretated as evidence for ID using classic ID arguments. If, as you say, it is highly improbable that a frameshift will generate something useful, then the prescence of over two hundred proteins created by frameshift is extraordinarily improbable - unless big G was controlling the shifts. On the other hand if the proteins were generated by some, as yet not understood method of moderating parents, then it might be a much more plausible mechanism.Mark Frank_{January 27, 2009
January
01
Jan
27
27
2009
11:05 AM
11
11
05
AM
PDT}

CJYman and others, Even if we do accept all the assumptions about "searching for a search," shouldn't we compute probabilities rather than talk about averages? I'll give an example following Dembski&Marks's construction on page 2 in the "search for a search." Let Omega={0,1} and search for 0. If we choose at random, the probability is 1/2 to find it. If we first choose a probability distribution on Omega at random, the probability to find 0 is still 1/2 (compute the expected value of p where p is uniform on [0,1]). That's the "conservation of uniformity" in this example. Now instead ask "What is the probability to do better than random search?" This is the probability to find a p that is greater than 1/2 and as p is uniform on [0,1], there is a 50% chance to beat random search. If we "search for a search" twice, we have a 75% chance to beat random search at least once. And so on. In other words, it is easy to beat random search by repeatedly randomly searching for searches. Now assume that Omega is huge but still finite, Omega={x1,...,xn} where we search for x1. [Note that x1 may represent an entire search string so finding x1 in one step may mean that we have found some target in a finite number of steps.] The probability to beat random search turns out to be about 37% [approximately = exp(-1)], regardless of n, and searching twice gives us about 60% chance to succeed. Conclusion: If we "randomly search for a search" at least twice, we are more likely than not to beat random search.Prof_P.Olofsson_{January 27, 2009
January
01
Jan
27
27
2009
09:56 AM
9
09
56
AM
PDT}

CJYman and others, The only logic I can see in how the "search for a search" paper supports ID would be to claim: "Either (a) the darwinian search algorithm was chosen according to the Kantorovich-Wasserstein distribution or (b) it was desiged." Note that the K-W distribution on M(Omega) depends on what metric you choose on Omega so there is no unique way to interpret "randomly choosing a search." Informal claims such as "it is as difficult to find a search as it is to find a target" are not supported by the paper unless you make a lot of arbitrary assumptions.Prof_P.Olofsson_{January 27, 2009
January
01
Jan
27
27
2009
09:42 AM
9
09
42
AM
PDT}

CJYman,
These papers were merely showing that the extent to which an evolving system increases probabilities is proportional to the improbability of finding that system (evolutionary algorithm) in the first place. The papers [along with the NFLT] explicitly detail the fact that in order to increase probability of generating a given pattern, a set of laws matching search space structure and search procedure, which is even more improbable, must first be generated.
If I may layman-ize: Darwinian processes are generally limited to searching for short range targets in local genome space. In order to search for long range targets a set of specific conditions relevant to the long term ("a set of laws matching search space structure and search procedure") must first be found as well. And a Darwinian search in nature is limited in scope to that which increases overall survivability, not functionality for merely functionalities sake as with engineering projects using GAs. Does that summarize correctly?Patrick_{January 27, 2009
January
01
Jan
27
27
2009
09:14 AM
9
09
14
AM
PDT}

sparc, You should hang out with Americans more often. You will pick up common sense and some humor. I know you have referred to us as dummkopfs but you read things too literally. My use of the term SFCSI was meant to show the absurdity of Mark Frank's objections and my guess you too if you deny FCSI. Do you deny the reasoning of FCSI or that DNA is information or a code? If so then you can join us as dummkopfs. For those who do not know, sparc is a biologist who is unable to provide any critical analysis on the evolution debate. And as I said earlier on this thread we need more like sparc here who only can contribute inane remarks when they disparage ID. Go Sparc.jerry_{January 27, 2009
January
01
Jan
27
27
2009
06:58 AM
6
06
58
AM
PDT}

JayM: "Thanks for the detailed reply. It appears that I do understand you correctly. In order for Dr. Dembski to present these two papers as support for ID, though, he (or someone) needs to make the links you hint at explicit." I see where you are coming from and I think that the reason he doesn't make the connection explicit is because it seems that these papers are more of a response to inflated claims about how evolutionary algorithms can account for biodiversity. These papers were merely showing that the extent to which an evolving system increases probabilities is proportional to the improbability of finding that system (evolutionary algorithm) in the first place. The papers [along with the NFLT] explicitly detail the fact that in order to increase probability of generating a given pattern, a set of laws matching search space structure and search procedure, which is even more improbable, must first be generated. Thus, evolution is not, as some have claimed, a free source of information -- information being an increase of probability. This increase of probability, labeled active information in the paper, is merely pushed back to higher and higher levels. What the paper does show is that this increase in probability seen in the operation of evolution still needs an explanation. Evolution is not an explanation of its own success at increasing probability anymore than an EA can create itself absent foresight. From reading the paper, we can see that we have a few options ... infinite regress of active info. which is akin to "turtles all the way down" and never truly explains increases in information (probability). That "explanation" merely explains away by handwaving. The other two options are merely chance and law, or foresighted systems. As I have already explained, no one has shown background noise (chance) and an arbitrary set of laws (set of laws with no regard for future results) to be up to the task of generating active information. If someone showed such an experiment, the math within these papers would be falsified, since they forbid such increases in probability. However, we do know through our own experience and through the creation of AI systems that foresight using systems (systems which model the future and generate targets) do exist and do increase probabilities of generating specified or pre-specified patterns. I have explicitly, yet briefly, outlined the ID connection in my post #122. JayM: "Unless one subscribes to the multiverse theory, it isn’t clear that the physical laws of this universe are the result of any kind of search (or search for search) process." The multiverse theory provides no more of an explanation for life and evolution than it provides an "explanation" for this conversation we are having. It is merely a chance of the gaps "explanation" and there is no criteria for when we can invoke it and when we can't. Can we invoke infinite chance to explain the results of all scientific tests? If so, then we would never discover anything since the proper explanation is that "the multiverse did it." Sure, using an infinite resource of chance comes in handy when trying to explain everything and anything, but is it the best explanation? This is where other ID argument come in, such as CSI. If a computer program based on only chance and law (absent previous foresight) won't produce CSI, why would an infinite amount of such programs be a better explanation rather than foresight, which is indeed characterized by its generation of CSI? As to the universe being the result of search, as long as it is one of any number of possibilities in "quantum chaos," then it is a search as defined by these two papers and as I previously explained in my post #121. At least our universe is confined to an extremely small set of possible mathematically described universes which even allow life and evolution. If, however, our universe is the only option, then no there is no search involved but this would raise other extremely intriguing questions. However, the papers would still show that any simulation of evolution requires the previous raising of probabilities before the evolutionary algorithm even starts. As we all know, intelligent humans raise the probabilities by introducing problem specific information into the behavior of the interaction between search algorithm and search space, by programming law and chance in such an improbable way as to solve a given problem in the future. That basic concept has been proven in the NFLT and extended in these two papers. Moreover, human intelligence uses its foresight to add problem specific information. IOW, all simulations of evolution require foresight so, barring this universe being the only option, on what grounds could someone say that the "real deal" requires no foresight as a scientific fact or that there is "no evidence for teleology within life" or that "ID is unscientific?"CJYman_{January 27, 2009
January
01
Jan
27
27
2009
06:34 AM
6
06
34
AM
PDT}

JayM: # 147 "This overstates the case somewhat. While some full proteins are species specific, most do have underlying partial homologies. It is those that are used as evidence for common descent." And so? Even if you group the proteins which have strong homologies, you still have myriads of separated islands. Who has said that an island must be made of only one protein? Even if myoglobin is present in many living beings, it is still completely different form insulin, or from c-myc. And who has ever criticized the concept of common descent? I accept it. But what has common descent to do with traversing the ocean of possibilities? Let's take the example of myoglobin. At some point in natural history, this protein appears. It is about 154 aminoacids long. Where did it come from? The fact that, after its appearance, it is maintained in most species means only that the function has been exploited in most species. But the traversing is necessary to get to that function, the first time. That's why I said that djmullen's statement that DNA replication was not a search was suspiciously trivial. Maybe he (and you) have not well considered that the search happens before, when you have to find the information, and not when you simply copy or transmit it? "I’m afraid that “extremely varied” isn’t supported by the empirical evidence." As an answer, I quote again from my previous posts: from my post # 52: “The proteins we do know (and we know a lot of them) are really interspersed in the search space, in myriads of different and distant “islands” of functionality. You don’t have to take my word for that. It’s not an abstract and mathematical argument. We know protein sequences. Just look at them. Go, for example, to the SCOP site, and just look at the hyerarchical classification o protein structures: classes (7), folds (1086), superfamilies (1777), families (3464). Then, spend a little time, as I have done, taking a couple of random different proteins from two different classes, or even from the same superfamily, and go to the BLAST site and try to blast them one against the other, and see how much “similarity” you find: you will probably find none. And if you BLAST a single protein against all those known, you will probably find similarities only with proteins of the same kind, if not with the same protein in different species. Sometimes, partial similarities are due to common domains for common functions, but even that leaves anyway enormous differences in term of aminoacid sequence.” from my post # 118: "In the genome of E. coli, a well known bacterium, we can find at least 2018 proteins described, grouped in 525 groups of similar proteins. The biggest group includes 52 proteins, while the smallest groups include only two proteins each, and there are 291 of them." Do you still think that "extremely varied" is not supported by the empirical evidence? What empirical evidence are you suggesting? "but the underlying amino acid sequences are not all over the map as you imply." That is simply wrong. They are. Go back to the two proteins I took at random from the E. coli genome. They are more than 700 aminoacids long, each one of them. In the best possible alignments, four different ones, you can find only 40 identities (not consecutive). Theose two proteins are "all over the map". And the same can be said if you take any other pair of proteins from different groups in the E. coli genome, or from different families or superfamilies in the SCOP classification. And even proteins with homologies are very different one from the other. Homologies are partial. Some are very partial. Choosing a conserved part of the c-myc sequence (a very important transcritpion factor), i have performed a blastp serach with a sequence of only 7 consecutively conserved aminoacids. 7 aminoacids is not much, they correspond to a search space of "only" 10^9. The search was done against the whole database of known proteins, and guess what? Only c-myc molecules had that exact sequence in their primary structure. You had to drop at least one aminoacid to find that sequence in other kinds of proteins. Can you see how terribly specific even a small sequence of aminoacids can be? Just consider that the specific immunological response in higher animals and humans is targeted against very small aminoacid sequences, the epitopes, whose length is usually less than 10 aminoacids, and yet those small sequences are so specific that they are the basis for our defenses. So, you can see that protein sequences are at the same time: a) extremely varied and interspersed on the whole map of possible sequences b) extremely specific, with very short sequences representing often a definite signature for one protein. "Again, we in the ID camp need to be very careful to address the real positions held by ID opponents" r That's exactly what I am trying to be. "rather than strawmen that not only can be easily knocked down but that can also be used to distract endlessly from our core message." I don't think I am using any strawmen. You can see that I am very precise and explicit in my arguments and in my answers, trying to address exactly what others, including you, are saying. If sometimes I misunderstand what the other is saying, I am ready to apologize.gpuccio_{January 27, 2009
January
01
Jan
27
27
2009
06:30 AM
6
06
30
AM
PDT}

JayM: # 146 "The fact that most evolutionary mechanisms result in tiny changes between generations is not just something that djmullen is “affirming”, it is a repeated empirical observation." What do you mean? Obviously a sinle point mutation changes amnly one nucleotide, but if it is a frameshft or a stop mutation a whole protein can change. Deletions and inversions can cause great changes in one signle step. The problem is that all those steps are random, and that's why a frameshift mutation , which is the equivalent of a blind long leap into the ocean, has practically no hope to find an "island". I have the impression that you and djmullen are maybe suggesting that a change can be "small" in regard to the whole genome. But we have not to reason about the whole genome. That is senseless. Our unit of reasonong must remain the single protein coding gene. There are single point mutations which are incompatible with life. Much more difficult would be to find a single point mutation which gives a reproductive advantage to a complex organism. Empirical observation of "useful" mutations is limited to microevolutionary events, like antibiotic resistance and even nylonase. In those observed events, the edge of evolution seems to be, at present, about two coordinated useful aminoacid substitution. That is very little to explain how you can get to a new protein of 700 aminoacids. Therefore, there is no empirical observation of mechanisms which can eliminate the need for a blind search in the ocean of sequence possibilities. "but even hinting that djmullen’s characterization is inaccurate suggests a lack of understanding that can be exploited by ID opponents. We need to be careful in our public statements." djmullen's characterization is inaccurate. I take full responsibility for this statement. And I think I am very careful in my public statements (which, obviously, does not mean that I cannot be wrong, like anybody else). "The obvious objection to this suggestion is that we’re being asked to prove a negative." We have nothing to prove. What darwinists are assuming is obviously incredible. It's them that have to provide some example or model to make it at least debatable. Otherwise, they have no theorym they have no model, and especially they have no empirical support. Which is exactly the case. "This would be a true vindication of The Edge of Evolution." The edge of evolution is already vindicated. But, obviously, we can do even better, and we will.gpuccio_{January 27, 2009
January
01
Jan
27
27
2009
05:50 AM
5
05
50
AM
PDT}

gpuccio: Obviously I'm not an expert in biology like you, and I'm assuming that all the claims you make about these islands of functional proteins are based on published research but it is still hard to see how two research papers about the mathematics of search algorithms, which don't explain why they are relevant or applicable to biology, are actually applicable to biology, in fact as a comp scientist I found it hard to see how the first paper could be applied to any real world situation given the way the authors seem to misunderstand some of these algorithms. What is needed is for an expert in the field like you to take all this evidence on the awkwardness of these protein search spaces and the evidence presented by Dembski and Marks, write it up as a paper and submit it to a biology journal. I doubt it would get published because of the conspiracy but you would still be able to post it on the web as evidence of the way these darwinians are suppressing the truth.Laminar_{January 27, 2009
January
01
Jan
27
27
2009
05:36 AM
5
05
36
AM
PDT}

gpuccio @144
2) All the proteins we can observe are extremely varied both as primary sequence and as function. Many of them are even species specific.
This overstates the case somewhat. While some full proteins are species specific, most do have underlying partial homologies. It is those that are used as evidence for common descent. I'm afraid that "extremely varied" isn't supported by the empirical evidence. There is certainly enough difference to make one question how non-intelligent mechanisms could be the cause, but the underlying amino acid sequences are not all over the map as you imply. Again, we in the ID camp need to be very careful to address the real positions held by ID opponents rather than strawmen that not only can be easily knocked down but that can also be used to distract endlessly from our core message. JJJayM_{January 27, 2009
January
01
Jan
27
27
2009
05:05 AM
5
05
05
AM
PDT}

gpuccio @144
If you go on affirming that evolution “mutates a very very tiny percentage of the parental DNA”, then you have to explain why proteins are so different and interspersed in the ocean of sequences.
The fact that most evolutionary mechanisms result in tiny changes between generations is not just something that djmullen is "affirming", it is a repeated empirical observation. Even "large" changes like duplication or frameshifts don't change the genome significantly in the next generation, although they can provide more variety in subsequent generations. There is much that ID proponents can legitimately argue against MET, but even hinting that djmullen's characterization is inaccurate suggests a lack of understanding that can be exploited by ID opponents. We need to be careful in our public statements.
Or show that in all cases bridges exist bewteen the islands of function where a selectable and growing function is maintained in all the intermediates, and the jumps between one intermediate and the others is always tiny enough that no important search has to be performed. I really doubt that anyone has ever done that, or will ever do that, even for one single important case (I mean, for one case which is not simply a microevolutionary, insulated case of one-two aminoacids substitution with immediate and definitive selection). Nobody can build such a model for a very simple reason: it does not exist.
I would turn this around and suggest that it is an excellent avenue for ID research. If it can be shown that no path exists, that would destroy MET. The obvious objection to this suggestion is that we're being asked to prove a negative. The problem is not, however, as intractable as that. There are a limited number of mechanisms in MET. There are a limited number of base pairs and amino acids. There is at least some historical record of what paths were actually taken. It would be difficult, but by no means impossible to show that a particular protein could not have arisen by MET mechanisms. This would be a true vindication of The Edge of Evolution. JJJayM_{January 27, 2009
January
01
Jan
27
27
2009
04:57 AM
4
04
57
AM
PDT}

CJYman @138

Please correct me if I’m misunderstanding you, but it seems that you’re saying that Dr. Dembski’s papers provide support for the idea that the universe is designed, but have nothing to say about the likelihood that the mechanisms of MET can account for all the biological diversity we see.”
They provide evidence that life and evolution are designed. They show that in order for a system (including evolution) to increase the probabilities of discovering a function/pattern or solve a problem, the higher order system which causes that evolution must be at least as improbable.
Thanks for the detailed reply. It appears that I do understand you correctly. In order for Dr. Dembski to present these two papers as support for ID, though, he (or someone) needs to make the links you hint at explicit. Unless one subscribes to the multiverse theory, it isn't clear that the physical laws of this universe are the result of any kind of search (or search for search) process.
So, as a foundational explanation, we need to look for something which can increase probabilities and foresighted systems are capable of that, however background noise and arbitrary collections of laws (set of laws created without any consideration for future results) are not up to that challenge.
That's what we have to prove rather than assert, though. MET mechanisms demonstrably have some ability to allow biological organisms to adapt within some limits, so we can't make a blanket statement that foresight is absolutely required. We need to find out what the limits are and prove them both mathematically and experimentally before we can claim to have positive evidence for ID theory. I hope that Dr. Dembski's papers are eventually seen as one small step in that direction. JJJayM_{January 27, 2009
January
01
Jan
27
27
2009
04:42 AM
4
04
42
AM
PDT}

djmullen (143): if you have followed all the previous discussions about prtoeins, you should already have the answers. To sum up: 1) Proteins are the essential material on which evolution has to work. 2) All the proteins we can observe are extremely varied both as primary sequence and as function. Many of them are even species specific. 3) The ocean of possible proetin sequences is, indeed, an ocean, and a very big one. Functional proteins are located in myriads of small separated islands in that ocean, as can be easily verified comparing their primary sequences, which are well known in great detail for a lot of them (just think of my example of two real proteins, completely different one from the other, with two different functions and 3D structure, and a conbinatorial space of more than 20^700). 4) If you believe that all the existing proteome is derived from some original ancestor, with a much smaller proteome and simpler functions, and by unguided darwinian means, that implies that the ocean of possible sequences has been traversed myriad of times in the course of natural history, and even more amazingly at OOL, in some mysterious prebiotic setting. 5) That's why Dembski and Mark's work about random searches is absolutely relevant, if you want to defend darwinian theory. I am not going into the details of their work, but relevant to biology it certainly is. 6) If you go on affirming that evolution "mutates a very very tiny percentage of the parental DNA", then you have to explain why proteins are so different and interspersed in the ocean of sequences. Or show that in all cases bridges exist bewteen the islands of function where a selectable and growing function is maintained in all the intermediates, and the jumps between one intermediate and the others is always tiny enough that no important search has to be performed. I really doubt that anyone has ever done that, or will ever do that, even for one single important case (I mean, for one case which is not simply a microevolutionary, insulated case of one-two aminoacids substitution with immediate and definitive selection). Nobody can build such a model for a very simple reason: it does not exist. And, to save darwinian theory form the paradise of inconsistent fantasies, that should be necessarily done not for one, but for all known cases of different proteins with different functions. I hope that answers more explicitly your questions.gpuccio_{January 27, 2009
January
01
Jan
27
27
2009
03:23 AM
3
03
23
AM
PDT}

gpuccio: Although proteins are interesting, my main concern is with the subject of this thread, the Dembski and Marks papers and how they relate to evolution. As you say, when copying DNA, "...there is no search necessary." You even say this fact is so trivial, I don't even have to state it. So I ask again, what do Dembski and Marks' papers on searching through the vast search space of a particular genome have to do with evolution? So far as I know, evolution never does anything like that. It either does a direct copy of the parental DNA, which means no search at all, or it mutates a very very tiny percentage of the parental DNA which means it "searches" an area of the genome very very close to the known-good genome of the parents. I don't see any relevance of the Dembski-Marks papers to evolution at all.djmullen_{January 27, 2009
January
01
Jan
27
27
2009
12:33 AM
12
12
33
AM
PDT}

Mark (# 136): I quoted that paper just to show you an example of how many proteins exist in the proteome which are specific to one single species. That means that they show no homology to proteins in other species. Even in the category of PDFs, they were citing 17,554 species specific proteins! That seems really a lot to me. That does not mean, obviously, that all the remaining proteins are conserved in all the species. It just means that they show some homology (certainly of various level and significance) to "some" other proteins in the tested genomes. In the interesting article linked in Paul Nelson's recent thread about the tree of life, we find the following interesting statement: "The battle came to a head in 2006. In an ambitious study, a team led by Peer Bork of the European Molecular Biology Laboratory in Heidelberg, Germany, examined 191 sequenced genomes from all three domains of life - bacteria, archaea and eukaryotes (complex organisms with their genetic material packaged in a nucleus) - and identified 31 genes that all the species possessed and which showed no signs of ever having been horizontally transferred." 31 genes. That's remarkable, isn't it? And about homology, please remember that the assumption that any level of homology in protein sequences is evidence of unguided derivation is only another myth of darwinism. First of all, homology is tested against the null hypothesis of randomness. Therefore, it is perfectly obvious that we will find some homologies bewteen proteins which have a similar function, for the same reason that all motorcars have wheels. That's certainly statistically improbable in a random assemblage of pieces, but it is certainly expected in a collection of machines which are designed to move. And finally, even if homology can show derivation, in no way it proves "unguided" derivation. So, to sum up, I think that the origin of "de novo" proteins does remain a completely unsolved mystery for darwinian theory.gpuccio_{January 26, 2009
January
01
Jan
26
26
2009
11:06 PM
11
11
06
PM
PDT}

Mark: You are, as usual, a careful reader. # 135: I had obviously read with great attention the other papers mentioned in the Wikipedia page. I did not mention them explicitly in my post pusposefully, but I was referring exactly to them when I wrote: "The difficulty of obtaining even a single new protein, however, is evidently troubling darwinists ever more. That can explain the recent attempts to find an important role to frameshift mutations, which are the only hope for traditional darwinism to obtain radically new sequences in a “simple” way. So, although frameshift mutations are totaly unrealistic as a way to obtain functional information, darwinists are trying just the same to exploit them in their “models”, driven probably by sheer desperation." I hoped that someone mentioned them, so that we could deepen the discussion. Those papers are really the product of wishful thinking, or if we want of desperation. The only interesting thing in them is that the authors frankly aknowledge that, without the frameshift mutation echanisms, it would really be difficult for the model of darwinian evolution to explain the observed diversity of proteins. But the papers themselves are absolutely generic and speculative. They are of the kind: let's take all possible frameshifts in the human proteome, and just blast them against the existing genome, and see if we find some partial homologies somewhere more frequently than we would expect in a completely random system. In other words, they are only abstractly playing with thousands of sequences to give some support to an unbelievable assumption. The only real example of an observed frameshift mutation which would give a functional protein cited by them as a basis for their assumptions is, obviously, nylonase. And, obviously, nylonase is bogus. It is one of the most beautiful examples of a bogus assumption emphasized by darwinists for years as "truth" only because it seemed to give support to them against IDists. It is a very good example of what dogmatism can do to scientific reasoning. Very briefly, the fairy tale of nylonase's frameshift origin begins (and ends) with a paper from Susumu Ohno on PNAS in April 1984. It was a paper exactly of the kind of the more recent ones cited in wikipedia: highly abstract, and unsupported. Ohno has just observed that there was a "possible" ORF in the genome of the plasmid containing the gene for nylonase which "could" have been, in the past, a protein gene for a £never observed" ancestral protein which "could" have given origin of the observed gene for nylonase by a frameshift mutation. It is interesting that Ohno was suggesting that an observed protein (nylonase) "could" have originated from a hypothetical, and never observed, protein by an hypothetical, and statistically almost impossible, mechanism. That could have remained just an interesting but bizarre paper about an interesting, but almost certainly false, theory, if darwinists had not decided to make of it a favoured piece of propaganda, much like they have done with another useful fairy tale, Matzke's theory about the "evolution" of the flagellum. So, the frameshift origin of nylonase became "truth", and frameshift mutations became the tool of darwinian evolution to quickly and efficiently traverse, by statistical magic, that ocean of improbability which, at the same time, darwinists were swearing there was no need to traverse. And why not? They had empirical evidence for that! We had to wait more than twenty years to get rid of those myths, by means of the serious work of those serious researchers who, thanks God, still exist. And please, take notice that the supporting myth that nylonase was a "de novo" protein, emerged in a few decades by the wonderful mechanisms of darwinian evolution, was based on a quick and superficial statement, in the paper about its discovery, that it presented no obvious homology to a bunch of known proteins, and no other apparent function than digesting nylon. Both these statements were, obviously, false. Finally, I must thank Zachriel for having prompted me to review the current literature about nylonase by citing it on your blog. It took a little time, but it was really rewarding. Strangely, the important news about that very important acquisitions had not been boasted around by darwinists, and even we IDists had probably not noticed it. But I must give credit to Wikipedia for correctly citing it, even if with the understandable, but very awkward, attempt at covering the facts by the prompt citation of those other absolutely irrelevant papers.gpuccio_{January 26, 2009
January
01
Jan
26
26
2009
10:50 PM
10
10
50
PM
PDT}

systematic, functional complex, specified information or SFCSI
Do you think that it will really help to introduce yet another term? Currently we have IC (Behe) and CSI (Dembski) which seem well established in the ID community. In addition one occasionally finds FCSI (KairosFocus, Jerry, gpuccio) and FSIC (Gordon Mullings). And now Jerry just introduced SFCSI. Will a six letter abbrevation (e.g. IFSCSI) be the next step in the evolution of CSI?sparc_{January 26, 2009
January
01
Jan
26
26
2009
08:28 PM
8
08
28
PM
PDT}

"Other people have demonstrated that MET is most likely not up to the task of generating all the biological diversity and function which we see (wasn’t that the point of the Altenburg 16). However, that only means that we don’t know everything about how evolution works yet. Big Deal … there’s much to learn about the mechanisms." The MET is an evolving synthesis of processes that account for changes in a population over time. As such it is far different from what it was in the early 1940's when it was initially finalized. It changed dramatically with the discovery of the structure of DNA and all the multitude of processes surrounding micro biology in the next 40 years after Crick and Watson discovered the DNA structure. It is constantly changing today as things such as epigenetics are added. The two constants in all the various syntheses have been natural selection and naturalistic mechanisms for providing variation. Gradualism is even being thrown to the wolves but never natural selection and never naturalistic mechanisms of variation. Changes to the MET are not thought of as supporting ID. Read Hunter's new website. They now say that organism have the capability to evolve built in. Is this support for ID according to the evolutionary biologists? No, because this capability was selected after it developed naturally. Evolutionary biologists have more outs from a sticky situation than Jack Bauer is capable of executing in 24 hours.jerry_{January 26, 2009
January
01
Jan
26
26
2009
08:11 PM
8
08
11
PM
PDT}

JayM (#133): "This does not, as you touch on below, reflect the mechanisms that are part of modern evolutionary theory (MET). The search space for viable organisms is not uniform and the mechanisms are not random (although some do include a random component)." That's the point. Since evolution is not random, what allows it? The answer: a matching of search space to search procedure which is just as improbable as the effects which it produces. ie: if we wouldn't expect the chemical constituents which make up a human brain to randomly coalesce in someone's backyard pool, then we shouldn't expect background noise and an arbitrary set of laws to produce life and evolution. JayM: "If we’re going to argue that MET is an insufficient explanation for what we observe in the natural world, we need to address what evolutionary biologists actually say." But neither myself nor these two papers say that MET (+ all other evolutionary mechanisms yet to be discovered) is insufficient to explain the biology which we see around us. These papers take evolution as granted. These papers merely get to the foundational point of the debate -- "what causes life and evolution" and "is life and evolution possible absent previous foresight" and "will background noise and an arbitrary set of laws cause life and evolution." The rest of my post which you responded to outlines the significance of these questions and how these two papers plus ID Theory help to provide answers. JayM: "I just read the NFLT papers I could find on the web. I’m not sure I’d summarize quite the same way, but I think I get your point." I have also read through them and they stress the significance of matching search procedure to algorithm in order to achieve any significant results. They even conclude (from what I can remember) by discussing the importance of incorporating problem specific information into the behavior of the algorithm in order to solve the problem. JayM: "This seems to be a significant change to ID theory. Instead of attempting to demonstrate that the mechanisms of MET are insufficient, we’re now reduced to arguing that the physical laws that result in those mechanisms are unlikely." Actually, that isn't strictly an ID argument. Poking holes in evolution is not equal to an ID argument. Many non-IDers, such as James Shapiro, point to the insufficiency of MET to explain the diversity of life which we observe. I believe he is on the right track researching cellular non-random genetic engineering in which living organisms control their own evolution to an extent. ID Theory, by definition, is a search for patterns which signify previous intelligence. Once I actually began to understand ID Theory, I noticed that it is not an anti-evolution argument. In fact, these two papers begin to provide evidence that evolution is one of those patterns which signifies previous intelligence. CSI and IC are other patterns which also signify previous intelligence. JayM: "The cosmological argument is interesting, but it cedes all of biology to ID opponents." Not so. Life and evolution itself is evidence for a previous intelligence. Think about it a bit. There would be no cosmological argument if life didn't exist. There would be no patterns to explain and thus no cosmological ID. Biological ID merely recognizes that certain features of life (including the system of life itself and its evolving process) require previous intelligence and Cosmological ID provides evidence that this intelligence is fundamental to the laws of nature. JayM: "MET doesn’t have anything to say about how the mechanisms arose, it merely shows how they explain what we see." That's pretty much the point. So when you see your friendly prof for the public understanding of science foaming at the mouth attempting to say that evolution proves that there is no teleology within life or nature, then you know that he is way out of his element. ID Theory is interested in quantifying the patterns produced by intelligence and what causes life and evolution. JayM: "Please correct me if I’m misunderstanding you, but it seems that you’re saying that Dr. Dembski’s papers provide support for the idea that the universe is designed, but have nothing to say about the likelihood that the mechanisms of MET can account for all the biological diversity we see." They provide evidence that life and evolution are designed. They show that in order for a system (including evolution) to increase the probabilities of discovering a function/pattern or solve a problem, the higher order system which causes that evolution must be at least as improbable. So, as a foundational explanation, we need to look for something which can increase probabilities and foresighted systems are capable of that, however background noise and arbitrary collections of laws (set of laws created without any consideration for future results) are not up to that challenge. Other people have demonstrated that MET is most likely not up to the task of generating all the biological diversity and function which we see (wasn't that the point of the Altenburg 16). However, that only means that we don't know everything about how evolution works yet. Big Deal ... there's much to learn about the mechanisms.CJYman_{January 26, 2009
January
01
Jan
26
26
2009
05:55 PM
5
05
55
PM
PDT}

Mark Frank, Use data as a synonym for the word information. The molecule has a name so you can use that or better yet since we know we are talking about nucleotides, so use A, T, C, G. Thus the information in the DNA or some segment of the DNA is the molecule A, T, C or G or a string of these letters. It is simple as that. For a rock you can use the molecule at some coordinate and then proceed to list every other molecule at each coordinate point. I cannot imagine why anyone would want to do this but it theoretically could be done and when through you would have a very complex set of information with each data point an individual piece of information. Now each triplet of letters in the DNA or certain parts of the DNA is another piece of information. What is so hard about this. If you want to pursue some philosophically obscure path to analyze this example be my guess but the rest of us will proceed on. Now certain segments of this DNA can be thought of as a unit and that unit is another level of information. Think of letter, word and sentence. And these sentences have meaning and that meaning is a string of polymers of amino acids which people call proteins. It is all quite simple and straightforward. So when we say that DNA contains information and complex information it becomes very obvious what is meant but so far it may not be any different than the rock. But the DNA data goes further and relates or specifies something else while the data in the rock has just hit a stone wall and is just still a rock. And we can go further and look at look at the thing specified and we see that it has a function so the nucleotides in DNA is information, complex, specifies something else and this something else is functional. We might even take this to a new level above this and say that the functional element or protein or RNA then becomes part of a coordinated system and maybe we can call the data or information in DNA something like systematic, functional complex, specified information or SFCSI. We have now left the rock in a distant galaxy as we accelerated to light speed to get to where the data is leading us to. And who knows that if we can determine new levels we will have a longer abbreviation that we can put in our Mickey Mouse decoder ring to get the meaning of life. If you want to quibble over this, be my guess but I do not see the point of it. I assume that these papers of Dembski and Marks are looking at the likelihood a sequence of mutations can lead a particular DNA string from one functionally complex specified point to another that is not related to it. If that is not what it is about then maybe someone could enlighten us peons.jerry_{January 26, 2009
January
01
Jan
26
26
2009
04:06 PM
4
04
06
PM
PDT}

Gpuccio I then went on to read the paper you reference in #129. It seemed surprisingly easy to understand. Of course I may have got it wrong, but it appears to conclude that there is a high (92.5%) conservation of proteins among the PDF proteins - those with clearly defined function. This among organisms as diverse as yeast and mammals with a last common ancestor hundreds of millions of years ago. There is much lower conservation among POF proteins. But these are the protein for which the function is not understood. And a lot of them have quite distinctive characteristics - so they cannot be treated as like PDF proteins for which the function has yet to be discovered. Many of these POF proteins might do nothing at all. Those that do something we have no idea whether it is the whole string that is relevant or just a part of the protein that is active or how wide a range of other proteins could perform the same function. If these things are true then this would greatly lighten selective pressure and under Darwinian assumptions they would diversify very rapidly - because a wide range of changes would be viable. Am I failing to understand something?Mark Frank_{January 26, 2009
January
01
Jan
26
26
2009
02:26 PM
2
02
26
PM
PDT}

Gpuccio You wrote: So, although frameshift mutations are totaly unrealistic as a way to obtain functional information, darwinists are trying just the same to exploit them in their “models”, driven probably by sheer desperation. What a pity that the only empirical support for those theories comes from the old model of nylonase, and that that model has been recently shown to be false (if you don’t believe me, just go to the Wikipedia page about nylon eating bacteria, and read carefully the linked paper by Negoro et al, to see how the famous nylonase, for years boasted by darwinists as an example of new function created by a single frameshift mutation, is instead a classical example of microevolution, where the mutation of one or two aminoacids allows a new substrate affinity in an existing esterase). About the fourth paragraph in Wikipedia reads: A series of recent studies by a team led by Seiji Negoro of the University of Hyogo, Japan, suggest that in fact no frameshift mutation was involved in the evolution of the 6-aminohexanoic acid hydrolase.[1] However, many other genes have been discovered which did evolve by gene duplication followed by a frameshift mutation affecting at least part of the gene. A 2006 study found 470 examples in humans alone.[2] with suitable references. What gives?Mark Frank_{January 26, 2009
January
01
Jan
26
26
2009
02:06 PM
2
02
06
PM
PDT}

Mark: excuse me if I have not answered your previous post (I will do that as soon as possible), but as you see the discussion has shifted to other issues. So I answer first you 131. Yes, you are right: great changes take place at molecular level in natural history, and we don't know why ot how. We in ID are certain that those changes are guided, for the reasons you know, but for the rest we are as ignorant as the darwinists. It is perfectly possible that a new protein comes out of an initial gene duplication, as darwinists believe. That's what programmers often work when they want to modify an existing code. And then in some way the new code is superimposed. I don't believe too much in the importance of working with big strings, because that has many restraints. In many cases you would need to change individual nucleotides at the right place, and point mutation remains the most powerful way to do that. I suggest here, just for discussion, some possible mechanisms of guided programming of a new protein. They are purely hypothetical, but I will mention where there could be some corresponding model in biology: 1) Guided variation: a variation which is not random, but pseudorandom, and where special events are intelligently favoured (for instance, at quantum level). That could include both point mutation and other mechanisms, like deletion, inversion, and so on. Guided vairation could also be mediated through intermediate tools, like transposons, ERVs, and others. 2) Random targeted hypermutation: we have an example of that in antibody maturation, but that is realized by an algorithm embedded in the immune system, and is not destined to genetic transmission. However that mechanism works, but it has to be coupled to: 3) Intelligent selection. In other words, the results which are in the sense of the needed change must be kept by a guided, intelligent intervention, even if they are not yet truly functional. For instance, the change in function could be measured, even if minimal, and preserved, as it happens in antibody maturation when the affinity for the original antigen is in some way measured (probably by the antigen presenting cells) and determines the suppression or retention of the new clone. That is similar to what happens in GAs, and it could take place in the context of the individual biological being, and not through the much slower (and imaginary) process of NS. Or, if the designer already knows the solution, he could act as a direct oracle, maintaining only the variations which are in accord with the information to be implemented (a la weasel). In all cases, intelligent selection, as we know from the example of appropriate GAs, can realize very quickly what NS can never achieve. But, while guided variation, targeted random variation and intelligent selection are certainly possible mechanisms by which a designer can implement intelligent information in the genome, huge mysteries do remain. First of all, is the implementation of the information really gradual, like traditional darwinism supposes? In other words, does the designer act in a very slow and gradual way? It is possible, but there are a number of issues against that: first of all, obviously, natural history, which is almost certainly discontinuous, especially at three major points: OOL, and the ediacaran and cambrian explosions. And the whole fossil record, in general, does not favour continuity, as Gould had well understood. And finally, as I have said often, we cannot really think that change can come hrough one new gene, however powerful. Real change in a complex system requires a redefinition of multiple genes and parameters, of the regulation, of the procedures, and of many other things. It is not a case that even very similar species are sometimes very different at the genomic level. Moreover, we still don't know how and where regulations and procedures are really implemented in the genome (or in some other place). The difficulty of obtaining even a single new protein, however, is evidently troubling darwinists ever more. That can explain the recent attempts to find an important role to frameshift mutations, which are the only hope for traditional darwinism to obtain radically new sequences in a "simple" way. So, although frameshift mutations are totaly unrealistic as a way to obtain functional information, darwinists are trying just the same to exploit them in their "models", driven probably by sheer desperation. What a pity that the only empirical support for those theories comes from the old model of nylonase, and that that model has been recently shown to be false (if you don't believe me, just go to the Wikipedia page about nylon eating bacteria, and read carefully the linked paper by Negoro et al, to see how the famous nylonase, for years boasted by darwinists as an example of new function created by a single frameshift mutation, is instead a classical example of microevolution, where the mutation of one or two aminoacids allows a new substrate affinity in an existing esterase).gpuccio_{January 26, 2009
January
01
Jan
26
26
2009
01:43 PM
1
01
43
PM
PDT}

CJYman @121
What is the probability that a given pattern (measured in bits) will be generated by unguided (random) bit operations? That is simply calculated as the pattern’s probability — assuming a uniform search space.
This does not, as you touch on below, reflect the mechanisms that are part of modern evolutionary theory (MET). The search space for viable organisms is not uniform and the mechanisms are not random (although some do include a random component). If we're going to argue that MET is an insufficient explanation for what we observe in the natural world, we need to address what evolutionary biologists actually say.
Now, the point of these papers is to ask [and answer] the question: “how can the probability of generating a pattern be increased *and what is the probability of finding a way to increase the probability of finding that given pattern*?” Little bit of a tongue twister, but read it a couple times and it’ll make sense. Well, basically, there can exist a search procedure (bit flipping operation) which rejects some bit operations/flips and keeps others. However, if the search space is uniform, this does not increase the probability for finding a given pattern. What needs to happen is that the search procedure needs to be matched with the proper search space that will allow the filter to actually improve the probability for finding a given pattern. This has already been proven in the NFLT.
I just read the NFLT papers I could find on the web. I'm not sure I'd summarize quite the same way, but I think I get your point.
Dembski and Mark’s paper picks up from there. What is the probability of matching a search space to a search procedure in order to increase the probability of finding a given pattern? They have merely shown that the probability of finding that match can be no less (and apparently increases exponentially with every higher level search) than the probability of finding the given pattern within a uniform search space in the first place. So appealing to a non-uniform search space and an evolutionary filter does not provide a solution to increasing the probability, since the probability (information) is merely moved back a step to the probability of finding the set of laws and initial conditions which will provide a non-uniform search space and ratcheting filter to increase the probability of finding the given pattern.
This seems to be a significant change to ID theory. Instead of attempting to demonstrate that the mechanisms of MET are insufficient, we're now reduced to arguing that the physical laws that result in those mechanisms are unlikely. The cosmological argument is interesting, but it cedes all of biology to ID opponents.
Thus, according to the math provided by Dembski and Marks, it is just as improbable to find a given pattern (measured against a uniform search space) as it is to find the search procedure (and landscape) to increase the probability of finding that given pattern.
MET doesn't have anything to say about how the mechanisms arose, it merely shows how they explain what we see. Please correct me if I'm misunderstanding you, but it seems that you're saying that Dr. Dembski's papers provide support for the idea that the universe is designed, but have nothing to say about the likelihood that the mechanisms of MET can account for all the biological diversity we see. JJJayM_{January 26, 2009
January
01
Jan
26
26
2009
01:16 PM
1
01
16
PM
PDT}

gpuccio @129
New molecular functions arise in new species. New proteins, new protein cascades, new regulations. New organs, new tissues, new cells, There is a lot of novelty, wherever you look at, in biology.
Certainly. There is also a lot of commonality. The ratio between those two is one of the pieces of evidence for common descent. Organisms with a more recent common ancestor share more features than those who diverged earlier.
djmullen’s reasonings of “sweet places” and “no searchs” are pure imagination, and bear no relationship to biological reality. And you accept them for the same reason: you don’t know how things are.
With all due respect, my knowledge of biology is more than sufficient to recognize that djmullen's comments are spot on. It's very simple to understand just by looking at yourself and your parents. You do not share a genome with either. In fact, you have a number of mutations so that your genome is not even a proper subset of the combined genomes of your parents. Nonetheless, you live. You are a viable organism. That demonstrates beyond a shadow of a doubt that there is a "sweet spot" in genome space within which the evolutionary mechanisms that operate between one generation and the next work. This is why I have great difficulty seeing how Dr. Dembski's two papers are at all applicable to the ID issue. I could see how they might be applied to the origin of life problem, but evolutionary biology starts from a viable point and explores nearby points, most of which are also viable. Evolutionary mechanisms do not search the whole possible multidimensional genome space.
Again, that’s not your fault, or djmullens’ fault. It is really strange how many people believe that the difference between species are easily explained by some reshuffling of existing information.
I've been enjoying our conversation, but your patronizing tone in this post is unappealing. You seem to be conflating multiple topics. In the context of observed evolutionary mechanisms, Dr. Dembski's papers do not seem to be obviously supportive of ID. The fact that small changes to a viable genome result in other viable genomes, as shown by the differences between you and your parents, indicates that the viable regions in genome space are not uniformly distributed. That means that mathematical discussions of searching the whole space, however elegant they may be, do not apply in this domain. The broader question of just how far these mechanisms can go is, as you point out, quite interesting. The fact that there are limits at one scale does not invalidate djmullen's explanation at the scale he was discussing. JJJayM_{January 26, 2009
January
01
Jan
26
26
2009
01:05 PM
1
01
05
PM
PDT}

Gpuccio I am thinking about your post #129. I wish there were a biochemist watching this discussion but I have a logical query. If: a lot of proteins arise practically “de novo” at sone time in the course of natural history, even if they can persist after in many species. what is the mechanism that causes that to happen (whether it be guided or not)? It is presumably a rearrangement of DNA in the genome and we know the mechanisms involved: * point mutation - clearly not sufficent to produce a radically new protein * crossing over during meiosis - but you seem to be saying new proteins arise in asexual species as well as sexual It seems that what we are really talking about is: * insertions, deletions, and duplications. These operations could instantly create new proteins that appear radically different from their predecessors. But they would be working with strings of DNA that had proven effective before. The unit being rearranged is much larger than a single base pair or even a single amino acid. I don't have enough biochemistry to know if this makes sense but it seems worth pursuing.Mark Frank_{January 26, 2009
January
01
Jan
26
26
2009
11:37 AM
11
11
37
AM
PDT}

Sal Gal: I can only quote myself (#104): "I am saying only two things: a) The genetic code is a symbolic code, in the sense that the information stored in DNA as protein coding genes can only be retrieved by means of a complex system, the translation system, where exactly the same symbolic correspondences are embedded. And that correspondence is not in any way connected to biochemical laws, but only to a semantic connection between the stored information in DNA and the translating system in tRNAs. b) The information in protein coding genes is functional information, because it is perfectly apt to guide the synthesis of a perfectly functional protein. Please notice that the function is in no way present in the DNA sequence (DNA can never act as an enzyme), but arises only in the final protein, as a consequence of the information of the DNA. The possibility of errors in the process does not change anything of that." To be even more clear, I am saying: 1) UCU is translated to serine in practically all protein coding genes. 2) That happens for no biochemical property which connects the tri-nucleotide UCU to the aminoacid Serine, but only because in all those beings there is another molecule, a specific tRNA, which has an anticodon in a key position which can connect (through biochemical laws) to the UCU codon in mRNA, and that same molecule (the specific tRNA) has another, separated site, in its structure, which links the aminoacid Serine and attachs it to the growing protein sequence. 3) The results of all that is that Serine strangely appears at the right functional place in all existing functional proteins. 4) The same can be said for all the other 19 aminoacids, and for all the other codons in the genetic code (except the stop codons), where each aminoacid is "mounted" by the appropriate tRNA, recognizing the appropriate codon on mRNA. 5) I am curious about that, and would like to know why it is that way. I don't accept that as a law of nature. And I don't accept the explanations of darwinists (provided that they have ever offered one). 6) If you are not curious about that, or if deeper problems prevent you from asking yourself such trivial questions, I really don't know what else to say. In case you want to go on with the discussion, could you please explain, as clearly as possible, what you don't agree with in the above, very simple, points? And what in them is "reification"? Thank you. And again, I certainly admire the complexity and creativity in nature, and often I am really overwhelmed at its depth and beauty. I feel many different emotions and intuitions when I look at nature, both in a garden or in a biology text (although the garden is usually better). But what I really "wonder" at is the darwinist folly, and how many intelligent people can share such a theory and such arguments without any apparent doubt or second thought. That is a real wonderment for me. That I would bet against, in my best Bayesians moments, as absolutely improbable, if I did not see it happen every day.gpuccio_{January 26, 2009
January
01
Jan
26
26
2009
11:03 AM
11
11
03
AM
PDT}

JayM (#120): I am afraid that you are wrong in what you say, and again that derives form little familiarity with biology (which, obviously, is not your fault). The problem is that darwinist must have convinced people, more or less indirectly, of a lot of things that are absolutely false. The point is that what I say does not refer "only" to the problem of OOL. It is perfectly true also for the problem of evolution. Indeed, I could well have made my example with two proteins from different species. Let's be clear: some proteins can be found in meany species, at different levels of natural history, and remain very similar, or chance in a more or less limited range. But a lot of proteins arise practically "de novo" at sone time in the course of natural history, even if they can persist after in many species. Are you suggesting that practically all the essential proteins, and 3d structures of proteins, were already present in the mythical LUCA? Well, that's not the case. Almost any species has proteins which are exclusively found in that species, and have no homology with other known proteins. In a recent paper "What makes species unique? The contribution of proteins with obscure features". available online on Genome Biology, a comparison of "predicted proteomes derived from 10 different sequenced genomes, including budding and fission yeast, worm, fly, mosquito, Arabidopsis, rice, mouse, rat, and human" showed that "7.5% of the proteins with defined features (PDFs) were species specific (17,554 in total)" and "60% of the Proteins with obscure features (POFs) identified in these 10 proteomes (44,236 in total) were species specific". And "Approximately one-quarter of eukaryotic proteins are POFs". The simplest bacteria have approximately 500 proteins. E. Coli has more than 2000. Eukaryotes are much more complex. Do you believe that all the proteins in eukaryotes can be traced to those in simple bacteria, with only minor evolutional "play". No. That's not true. C. elegans, matbe the simplest multicellular being, a small worm made of only 1000 cells, has a proteome of about 20000 proteins, almost the same as humans. Do you believe they are the same? Or that all human proteins are very similar to those in the proteome of C. elegans? New molecular functions arise in new species. New proteins, new protein cascades, new regulations. New organs, new tissues, new cells, There is a lot of novelty, wherever you look at, in biology. djmullen's reasonings of "sweet places" and "no searchs" are pure imagination, and bear no relationship to biological reality. And you accept them for the same reason: you don't know how things are. Again, that's not your fault, or djmullens' fault. It is really strange how many people believe that the difference between species are easily explained by some reshuffling of existing information. Perhaps they read too much Dawkins (certainly not a good idea), and their best model is dog breeding. But sometimes I wonder how darwinists have succeeded in spreading so much disinformation about biology, and in hypnotizing people making them believe things that have no scientific basis.gpuccio_{January 26, 2009
January
01
Jan
26
26
2009
10:38 AM
10
10
38
AM
PDT}

Prev 1 … 3 4 5 6 7 … 10 Next

You must be logged in to post a comment.

Leave a Reply