Friday, August 1, 2008

The Design Matrix Revealed

This is the third in a series of posts regarding themes in Darwin’s Black Box and The Design Matrix.

All quotes in this post are from Chapter 10 of The Design Matrix.


The Design Matrix

Mike Gene’s book, The Design Matrix, is an interesting and enjoyable read which culminates in Chapter 10 where the Design Matrix (DM), a method to gauge the degree of design in a natural object, is described in detail.

The DM has four separate criteria for scoring*:

1. Analogy (A): The more similarities of biotic features to features known to be designed (engineered), the stronger the analogy.

2. Discontinuity (D): Non-teleological processes cannot explain biotic feature. When A is coupled with D, the design suspicion is strengthened.

3. Rationality (R): Does the biotic feature display function and “purpose”? IOW, would a competent engineer design it that way?

4. Foresight (F): Are there aspects to the biotic feature(s) that point towards thinking forward rather than immediate success? (Engineer with foresight vs. myopic tinkerer)

“The Design Matrix … works by taking the scoring along each criterion and fusing them together as a whole… In this way, the four criteria are treated independently as possible, yielding their own respective score. Then, the scores are simply averaged to give a final score…”

Each criterion is given a score by the user that ranges from -5 to +5 where a positive score leans towards a suspicion of design while a negative score suggests no design. For example, if the evidence either way is “strong”, then +/- 5 is scored; if moderate, +/- 3; if modest, +/-1. Zero represents a thoroughly ambiguous situation. Once all four criteria are scored, their average is taken: sum of A, D, R, and F divided by 4. The final score suggests whether a suspicion of design is warranted or not.

Mike Gene goes through several examples. While I’ll leave it to the reader to sift through the details, here is how Mike scored various objects, both natural and human designed (A, D, R, F, final score):

Pseudogene: -5, -5, -5, -3, -4.5
PCP Pathway: +1, -5, -4, 0, -2
Eye: +3, -3, +2, -2, 0
Genetic Code:+4, +2, +4, +2, +3
Book: +5, +5, +4, +3, +4.25
Car: +5, +5, +4, +4, +4.5

Thus, according to Mike’s scoring, the genetic code demonstrates a moderate suspicion of design, the pseudogene and the PCP pathway lean towards non-teleological explanations, and the eye is ambiguous. It should also be noted that the scores are subject to change should new evidence come forth.

Strengths and Weaknesses

“Again, we must be clear that the scoring is not objective. The Design Matrix is not intended to be a scientific instrument, …”

As I have surmised in previous posts, the DM is not an objective measure of design in nature. So what use does it have? Let’s look at its strengths to help us define its use.

“Combining the scores … happens to eliminate one of the most popular arguments against design – ‘god of the gaps.’ A Discontinuity score, by itself, is quite vulnerable to this complaint. But if the Discontinuity score is combined with three lines of positive indicators of design, the “gaps” complaint no longer applies.”

“Within the Matrix design is not inferred simply because there is a lack of evidence that something evolved. Instead, such considerations are simply one piece of the puzzle, where Discontinuity, combined with Analogy, Foresight, and Rationality give us a broader perspective with which to reach a tentative conclusion.”

With the DM, Mike Gene combines scores from four (mostly) independent lines of evidence into a single score that suggests whether teleological processes (design) should be suspected or not. This means that the negative evidence of IC can be augmented by positive evidence from A, R, and F, thus strengthening the design suspicion. However, the opposite can be true: a lack of positive evidence from A, R, and F weakens the suspicion of design raised by the negative evidence of IC.

The second strength of the DM is it forces each user to put their reasoning for scoring out in the open. Willy-nilly scoring won’t cut it. Each score has to be supported evidence. As in court cases, the more evidence, the stronger the case. Also, independent scoring from a separate user of the DM can point out both strengths and weaknesses in the original scoring.

Summary

“If you, the reader, still find yourself wanting independent evidence of a designer and needing some part of evolution to be disproved, you will have been disappointed.”

While the DM is not an objective measure of design in natural object, it can be a useful tool, helping to provide direction to those who wish to “follow the Rabbit**”. Suspicions of design can be strengthened and thus a direction for future research and experiments can be made clearer.


*These descriptions are brief. For a more detailed description, read Chapters 8 and 9 of The Design Matrix.
**The Rabbit and Duck theme is a favourite of Mike’s. It has its origins from this drawing, suggesting two people can look at the same thing yet see two different things. Mike defined the Rabbit as those who see design in nature and the Duck as those who see nature caused by non-teleological mechanisms.

Source:
Gene, M., The Design Matrix: A Consilience of Clues, 2007

Next: ID Research Themes

36 comments:

  1. "...the Design Matrix (DM), a method to gauge the degree of design in a natural object, is described in detail."

    and later,

    "The final score suggests whether a suspicion of design is warranted or not."

    Is it the degree of design or is it the degree of suspicion that is being scored?

    BTW, thanks for the introduction to Mike Gene's book. I'll try to read it soon because it makes references to engineering.

    Mike Gene deserves credit for distancing himself from the Intelligent Design Movement and for acknowledging that what he is doing is not science.

    He's not proposing any new models of how the world works; he's just philosophizing. More specifically, he's proposing a way to philosophize in a systematic fashion. Which is fine for the people who find that interesting. The rest of us will go on trying to build models.

    ReplyDelete
  2. I hasten to add that people can be interested both in philosophizing and in model-building. Carl Sagan, who you brought up before, is a good example. Mike Gene himself is another good example; it is my understanding that he is a scientist in his "day job." Even better, both of them are examples of people know the difference.

    ReplyDelete
  3. "Is it the degree of design or is it the degree of suspicion that is being scored?"

    I think the argument could be made for both. IMO, Mike Gene's intent for the Design Matrix was to gauge a specific design suspicion. However, when you read Chapter 10, there is a range from +1 to +5. It could also give one an idea of the degree of design involved.

    "He's not proposing any new models of how the world works; he's just philosophizing. More specifically, he's proposing a way to philosophize in a systematic fashion. Which is fine for the people who find that interesting. The rest of us will go on trying to build models."

    And yet, Mike has done more. He has laid the foundation for ID research (see this post). Let the experiments validate the philosophy.

    ReplyDelete
  4. The way I see it, the DM amounts to little more than raising to the fourth power the notion "it looks that way to Mike Gene, thus design". I believe if one approaches this a bit more objectively, one can see how tenuous the DM really is.

    For example, here is how I would rate the genetic code:

    Analogy: -2 (it's only from a very special perspective that the genetic code looks like anything designed. There is no reason to restrict our viewpoint to a single highly-biased one.)

    Discontinuity: -5 (the fact that the genetic code has a simple stereochemical basis makes this the only possible score)

    Rationality: 0 (see below)

    Foresight: 0 (see below)

    The bottom line - this is one of the better cases an ID proponent may make, and the results are solidly on the "not designed" side.

    More generally, as I've said on the ARN boards, it's apparent to me that nothing can really pass muster as looking designed. As has been shown on ARN, there can be no reliable analogy score; thus, on average, it will be zero. Given that all ID arguments about discontinuity are negative, the best score on this point is zero. Any positive evidence at all against discontinuity (such as linking sub-systems with other functionalities) moves the score down. Rationality is as subjective as analogy, and thus as unreliable. Foresight can at best be zero, since biology is at its core a yin/yang proposition.

    ReplyDelete
  5. Welcome to EE, Art.

    Interesting scoring of the genetic code. As I've stated in this post:

    "The second strength of the DM is it forces each user to put their reasoning for scoring out in the open. Willy-nilly scoring won’t cut it. Each score has to be supported evidence. As in court cases, the more evidence, the stronger the case."

    Forgive me for favouring Mike's scoring over yours, not just because of bias, but because Mike's evidence is better than yours.

    ReplyDelete
  6. Just so I don't reveal too much of Mike's book, his evidence is summarised on page 281 (actually, the paragraph starts on the last line of page 280).

    Mike goes into further detail of the analogy strength in Chapter 4.

    The discontinuity is strong since there are no plausible explanations for the naturalistic origins of the genetic code.

    Mike describes in further detail the rationality and foresight involved with the genetic code in Chapter 9.

    Thanks for putting your cards on the table, Art. I just don't think you have a good hand.

    ReplyDelete
  7. Thanks for the speedy reply.

    I am afraid that, when it comes to evidence, Mike Gene's perspective falls way short.

    For example, unless you can show me a language that has the property that the meaning of words is determined by (and only by) the medium on which the language is conveyed, then my score for analogy is much better than yours (or Mike Gene's).

    Moreover, until you can explain away the experimentally-observed fact that the genetic code has a simple stereochemical basis that clearly shows a continuity with other aspects of the RNA World, then my score for discontinuity takes precedence (as it is supported by experimental observation).

    My other comments about rationality and foresight are obviously true. It astonishes me that someone would question them, as they are grounded in the most basic of biological realities.

    BTW, I've debated Mike Gene for 8+ years, and he has never offered anything by way of experimental evidence to counter my criticisms. I'm bringing up my points here so that you can appreciate that there are serious problems and unanswered questions with the DM.

    Again, thanks for the welcome.

    ReplyDelete
  8. Art said:
    "Thanks for the speedy reply."

    Everyone's asleep. I've set the sprinkler timer. Clothes are washing. That means it blogging time, baby! ;)

    "My other comments about rationality and foresight are obviously true. It astonishes me that someone would question them, as they are grounded in the most basic of biological realities."

    So pretend it's not obvious to me, and pretend I don't want to sift through the many discussions at ARN. Actually, you don't have to pretend for that second one ;) Seriously, I am interested in your views of rationality and foresight, in contrast to Chapter 9 of the Design Matrix.

    "Again, thanks for the welcome."

    Everyone is welcome at EE. I have a no-ban, no-deleted-comment policy (although I have been tempted at times).

    ReplyDelete
  9. One more comment for you, Art. You appear to put some, if not a lot, of stock in the RNA World scenario. I take it you're not a fan of Robert Shapiro and his Scientific American article from February 2007 (here)?

    ReplyDelete
  10. Shapiro has no problems with the RNA World. He is just of the opinion that some sort of prebiotic metabolism preceded the RNA World (as opposed to the flip side of the coin, that primordial RNAs instigated the first metabolic systems).

    I agree with Shapiro on this account. But not perhaps as many might expect. I hope to update the following essay for my blog, but for now the link is the best I can do. As always, enjoy!

    http://www.arn.org/ubbthreads/showflat.php?Cat=0&Number=215607&an=0&page=78#Post215607

    ReplyDelete
  11. Rationality:

    This is a highly subjective concept. If one is going to try and quantify things, as in the DM, then one must try to look at the whole picture. In biology, and in this charged debate, any quantitation of rationality is going to home in (rather quickly) on zero (using the DM scale). For example, one may imagine some rationality in the genetic code (as in the idea that the code is deliberately engineered to buffer mutation, or facilitate evolution). But one must also factor in other things, such as the unnecessary susceptibility to nonsense mutations that is built into the genetic code. Or the existence of mutational mechanisms quite apart from C deamination that render the latter mechanism far less of a factor than some may imagine.

    Etc., etc., and so on. As I said, biology is a yin/yang proposition, and this inevitably will cause a highly subjective exercise (such as evaluating rationality) to come to zero (again, using the DM scale). It's unavoidable.

    ReplyDelete
  12. Hi JJS, I forgot to sign in as Bilbo, but that's who this is writing.

    It looks like I got you some free advertising at TT. I joined ARN in 2001. From what I remember, Mike Gene usually had the last word in his debates with Art. But perhaps my perspective is prejudiced.

    As to Analogy, I know next to nothing about computers, but aren't the codes used basically ones where "the meaning of words is determined by (and only by) the medium on which the language is conveyed"?

    As to the Discontinuity of the genetic code, Jules had a go around with Art at ISCID, and I don't think Art demonstrated continuity nearly as well as he seems to think.

    As to Rationality, Art needs to show that there is a way to avoid nonsense mutations, without destroying the ability to mutate at all. Further, Mike Gene has already written an essay on the question of how other mutations would affect C deamination.

    As to Foresight, if computer simulations are correct, then our genetic code is one of the most optimal there is, able to preserve design far into the future.

    ReplyDelete
  13. Art, I never put the Design Matrix forward as a "foolproof" way of finding design in nature, and neither did Mike Gene:

    "While the DM is not an objective measure of design in natural object, it can be a useful tool, helping to provide direction to those who wish to “follow the Rabbit**”
    JJS P.Eng. Last Paragraph. This post.

    “If you ... still find yourself wanting independent evidence of a designer and needing some part of evolution to be disproved, you will have been disappointed."
    Mike Gene, Chapter 10, Design Matrix

    ReplyDelete
  14. Welcome to EE, Bilbo.

    "It looks like I got you some free advertising at TT."

    You mean I went to the bank machine for nothing??? ;)

    As for being "Mike Gene's interpreter", how much does that pay? ;)

    Thanks for posting my review at TT. It's been the high-point of my very short blogging "career" :)

    ReplyDelete
  15. Hi Bilbo,

    Some very brief comments for this evening.

    "As to Analogy, I know next to nothing about computers, but aren't the codes used basically ones where "the meaning of words is determined by (and only by) the medium on which the language is conveyed"?"

    Um, what I meant was that codes and languages are arbitrary constructs. There is no physical connection between the letters "cat" and a feline. The same is not true for the genetic code. Triplets code for amino acids because they are connected (historically) with those amino acids. Thus, when it comes to analogy, the genetic code is a solid negative.

    "As to the Discontinuity of the genetic code, Jules had a go around with Art at ISCID, and I don't think Art demonstrated continuity nearly as well as he seems to think."

    I don't recall Jules ever having commented on the work of Yarus and coworkers. Care to help me refresh my memory?

    "As to Rationality, Art needs to show that there is a way to avoid nonsense mutations, without destroying the ability to mutate at all."

    Not at all. I need only point out that one stop codon is all that is needed. Three stop codons = much more "opportunity" for nonsense mutation.

    " Further, Mike Gene has already written an essay on the question of how other mutations would affect C deamination."

    I missed this as well. Got a pointer?

    "As to Foresight, if computer simulations are correct, then our genetic code is one of the most optimal there is, able to preserve design far into the future."

    The genetic code is optimal only to the extent that evolution optimized it. It actually has not preserved the designs in which it first appeared.

    ReplyDelete
  16. Art, here is the link to the discussion you had with Jules (me, of course) at iscid(don't know how to do that blue thingy without help): http://www.iscid.org/ubb/ultimatebb.php?ubb=get_topic;f=6;t=000606#000001

    It seemed to involve Yarus's findings.

    ReplyDelete
  17. Art said: Um, what I meant was that codes and languages are arbitrary constructs. There is no physical connection between the letters "cat" and a feline. The same is not true for the genetic code. Triplets code for amino acids because they are connected (historically) with those amino acids. Thus, when it comes to analogy, the genetic code is a solid negative.

    Which is why I mentioned computer codes. We come up with a code, then all the instructions that determine how the computer "interprets" the code. There is no conscious mind in the computer that connects the code to what to do with it, in contrast to our minds that connect "cat" to a certain animal. These instructions in the computer are the same sort of thing like the enzymes (proteins) that connect the proper tRNA to the proper amino acid.

    So, yes, the genetic code is very much like the kind of codes we create for non-conscious machines.

    Art: I need only point out that one stop codon is all that is needed. Three stop codons = much more "opportunity" for nonsense mutation.

    Ah, interesting. So here is where an ID researcher would have an opportunity to do research: Why would a designer use 3 stop codons, instead of 1? Or do we already know the advantage of redundant stop codons?

    Art: The genetic code is optimal only to the extent that evolution optimized it. It actually has not preserved the designs in which it first appeared.

    Or has it not preserved them because they were never there? In his book, Mike Gene argues that we have found organisms where the code has evolved, even though it is not as optimal. Yet organisms that use it manage to survive in their own little niches. If there were previous codes, it seems logical to think that there would be little niches where they were preserved.

    ReplyDelete
  18. Mike Gene's second essay on C deamination:

    http://idthink.net/biot/deam2/index.html

    ReplyDelete
  19. Hi Bilbo,

    Thanks for the pointers. FYI, Mike Gene’s C deamination pt2 avoids the matter of additional mutational mechanisms. It also avoids some other issues that were brought up on the ARN boards. All in all, his case is pretty slim.

    As for Jules’ contributions, I think your statements call into serious question the application of the DM for discovery and understanding. On the ISCID boards, we learned of two solid, incontrovertible, and as yet un-refuted pieces of data that link the genetic code with the RNA World. (Jules never offered anything to counter the facts and their implications, and neither I daresay has any other ID proponent.) You remain unmoved, and I suspect that you would support jjs’ assignment of a value of +2 for the “discontinuity” of the genetic code. IMO, this really renders the DM as pointless. Here’s why:

    I would think that the default value for any of the DM categories would be 0, and I would think that any movement away from 0 would require some sort of experimental evidence. I have pointed out two as yet unchallenged studies that support the idea of continuity. As of the moment, we don’t have a “singularity to the human genome” demonstration in the lab, but we still have solid data that is not consistent with discontinuity. This is why I gave a value of -2.

    jjs (and all other ID proponents) offer no experimental data to support the notion of discontinuity – none whatsoever. So, from the perspective of experiment, the case is crystal clear, and as I have stated.

    In this light, and given that I am rather sure that you and jjs will argue for a positive value here, it seems to me that the use of the DM is quite apart from empirical study. Y’all don’t seem to be interested in actual data. What this leaves us is evidence-free suspicion. But that’s where we were before the DM – unsupported suspicion, wishful thinking, and the like. The DM doesn’t add anything to the toolkit, and indeed seems more than ever to be an artificial multiplier of “it looks that way to me, thus design”.

    ReplyDelete
  20. Art wrote: Thanks for the pointers. FYI, Mike Gene’s C deamination pt2 avoids the matter of additional mutational mechanisms. It also avoids some other issues that were brought up on the ARN boards. All in all, his case is pretty slim.

    Could you "point" me in the right direction for specific objections?

    As for Jules’ contributions, I think your statements call into serious question the application of the DM for discovery and understanding. On the ISCID boards, we learned of two solid, incontrovertible, and as yet un-refuted pieces of data that link the genetic code with the RNA World.

    I re-read the thread, and I think I saw one solid piece of data: 8 amino acids display affinity for their codons or anti-codons. I missed the second piece of data.

    (Jules never offered anything to counter the facts and their implications, and neither I daresay has any other ID proponent.)

    True, but I was just trying to understand the data at that time. Let me offer an ID perspective, now. A knowledgeable designer, realizing that there was an affinity between some of the amino acids and their codons or anti-codons, designed the tRNA synthases so that the amino acids would bind to tRNAs with those codons, thus avoiding potential problems of interference by binding to the wrong tRNAs.

    You remain unmoved, and I suspect that you would support jjs’ assignment of a value of +2 for the “discontinuity” of the genetic code.

    Yeah, showing that 8 amino acids bind to their codons or anti-codons didn't "move" me much, since I wanted to know why they would bind to the stems instead. Speculation that in an RNA world things were different and would explain all of that didn't "move" me much either. Have any of those details been filled in? What I score for Discontinuity doesn't matter much, since I'm just part of the peanut gallery. But so far, looking from the sidelines, +2 looks about right.


    IMO, this really renders the DM as pointless. Here’s why:

    I would think that the default value for any of the DM categories would be 0, and I would think that any movement away from 0 would require some sort of experimental evidence. I have pointed out two as yet unchallenged studies that support the idea of continuity. As of the moment, we don’t have a “singularity to the human genome” demonstration in the lab, but we still have solid data that is not consistent with discontinuity. This is why I gave a value of -2.


    I agree that a default value of 0 is a good place to start. I think the data that shows some sort of continuity is weak, but should be included in our score. That's why +2 instead of higher. But I'll go down to +1 if you go up to -1. Deal?

    jjs (and all other ID proponents) offer no experimental data to support the notion of discontinuity – none whatsoever. So, from the perspective of experiment, the case is crystal clear, and as I have stated.

    Since most informed commenters (regardless of their views on ID) on the genetic code seem to think that it is arbitrary, I'm guessing that it was based on some sort of experimental data. And showing that 8 amino acids have an affinity for their codons doesn't make the case crystal clear.

    In this light, and given that I am rather sure that you and jjs will argue for a positive value here, it seems to me that the use of the DM is quite apart from empirical study. Y’all don’t seem to be interested in actual data. What this leaves us is evidence-free suspicion. But that’s where we were before the DM – unsupported suspicion, wishful thinking, and the like. The DM doesn’t add anything to the toolkit, and indeed seems more than ever to be an artificial multiplier of “it looks that way to me, thus design”.

    The part about not being interested in actual data is what really hurts, Art. And I don't think you're giving Mike Gene enough credit* for expanding "it looks designed," into the four categories of Analogy, Discontinuity, Rationality, and Foresight. Even high scores in Analogy and Discontinutiy do not guarantee an overall high score for Design.

    *You of all people, the only ID critic included in his list of acknowledgements. Oh just thinking about it causes so much heartache. Take your football and begone!

    ReplyDelete
  21. Mike Gene just wrote up something on the advantages of 3 stop codons:

    http://www.thedesignmatrix.com/content/complementing-the-cytosine-deamination-story/#more-168

    ReplyDelete
  22. Seriously, Bilbo. How can you assert a positive value for discontinuity in the case of the genetic code when the only data - I repeat, the ONLY data - point to continuity.

    You show us quite clearly how IDists care nary an iota for data. If you were to go where the experiments lead, you would have to assign a negative value.

    I would be interested in the list of "informed commenters" who have reviewed the work of Yarus et al. and offered positive experimental evidence that is not in agreement with his work.

    (I know I'm asking a lot, but I am not inclined to equate the off-the-cuff comments of some IDists with the solid and compelling experimental results I am discussing. I want names and pointers, so I can see the specifics as far as commentary about the stereochemical basis of the genetic code.)

    ReplyDelete
  23. Mike Gene's blog entry has nothing to do with the matter I mentioned.

    ReplyDelete
  24. "True, but I was just trying to understand the data at that time. Let me offer an ID perspective, now. A knowledgeable designer, realizing that there was an affinity between some of the amino acids and their codons or anti-codons, designed the tRNA synthases so that the amino acids would bind to tRNAs with those codons, thus avoiding potential problems of interference by binding to the wrong tRNAs."

    That's just an unsupported and untestable assertion. I can as easily and as justifiably claim that a knowledgeable designer would have nothing to do with any of this, and thus that life as we know cannot be designed.

    I will need some time to dig up the reference I cryptically refrred to above - the second item of positive support for continuity. The study I have mentioned elsewhere showed how a part of the tRNA actually participated in the reaction catalyzed by tRNA synthetases. That's a bang-up piece of data that supports a very interesting hypothesis that flows from the RNA World scenario.

    ReplyDelete
  25. "And I don't think you're giving Mike Gene enough credit* for expanding "it looks designed," into the four categories of Analogy, Discontinuity, Rationality, and Foresight. Even high scores in Analogy and Discontinutiy do not guarantee an overall high score for Design."

    My main problem with the DM (as I'm reading it here) is that it seems to be a needless multiplier. If one is going to see an analogy, then rationality and foresight, at the very least, are going to be foregone conclusions. The same will usually be true for discontinuity (a property that is folded integrally into most analogies in the first place).

    "*You of all people, the only ID critic included in his list of acknowledgements. Oh just thinking about it causes so much heartache. Take your football and begone!"

    Football? You been talkin' to Matty somewhere?

    In any case, this is the second mention I've seen made of this. Seeing as Mike Gene thinks I accused him of child abuse and poor parenting, and that I'm some terrible bullying ogre, I sort of suspect that people are confusing me with someone else. I'm pretty sure he would not mention such a monster in any positive sense.

    ReplyDelete
  26. Hi Art, I'm running out of time (so many blogs, so little time). But I really miss Matty. Don't you?

    ReplyDelete
  27. From The Design Matrix, p.xiv, : "The Internet can be an extremely hostile environment and their support with stimulating questions, commentary, and/or kind words has been greatly appreciated. These people include...Art...".

    There, now don't you feel ashamed of yourself?

    ReplyDelete
  28. Art: Seriously, Bilbo. How can you assert a positive value for discontinuity in the case of the genetic code when the only data - I repeat, the ONLY data - point to continuity.

    Are we still talking about only 8 (out of 20) amino acids? And the catalyzing activity of tRNA that your going to reference? I'll agree that it supports continuity. The question is, how much does it support it? And it looks to me like the answer is, not a lot. But I'm out of time. Libraries closed on Sundays. Be back Monday.

    ReplyDelete
  29. Oops, forgot to sign in as Bilbo, again, but that's who this is.

    Art: Seriously, Bilbo. How can you assert a positive value for discontinuity in the case of the genetic code when the only data - I repeat, the ONLY data - point to continuity.

    I found another library. Data for discontinuity: Apparently 12 amino acids show no affinity for the (anti)codons. No remnants of an earlier code found. At best, highly speculative scenarios of how the code could have evolved from an RNA world.

    You show us quite clearly how IDists care nary an iota for data.

    Once again, ouch. I'm hurt that you have such a low opinion of me. I think I've earned better. As to Mike Gene, from his book, p. 281:
    "When it comes to Discontinuity, we must pay tribute to the fact that there are dozens, perhaps hundreds, of papers where scientists have speculated about the evolutionary origin of the code and gathered circumstancial evidence to support their respective cases. Neverthelss, the various explanations remain largely in the realm of speculation....the code is universal, lacking any trace of simpler precursors or permutations spawned from such ancient precursor states....if the code was indeed spawned from non-teleological forces acting over great spans of time, we would predict the existence of such permutations someplace in the biological world. That they do not exist argues against such an origin. As a result of the above considerations, I give the code a Discontinuity score of 2." If it hadn't been for the hundreds of papers by scientists speculating ona non-teleological origin, Mike probably would have given it a higher score.

    If you were to go where the experiments lead, you would have to assign a negative value.

    I think I've explained why I wouldn't. But I understand why you do, and I don't think you are entirely unjustified. I do suggest a -1, instead of a -2.

    I would be interested in the list of "informed commenters" who have reviewed the work of Yarus et al. and offered positive experimental evidence that is not in agreement with his work.

    I've only read Franklin Harold's book, The Way of the Cell, and I doubt he was familiar with Yarus's work. It could be that Harold would have changed his mind. There's also Maynard Smith and Stuart Kaufmann. Again, they might not have known about Yarus, but Kaufmann makes the point that the arbitrariness is apparent from the fact that the amino acids bind to the stem of the tRNAs, and not the anticodons.

    Art said...
    Mike Gene's blog entry has nothing to do with the matter I mentioned.

    Well, if I recall, you thought the 3 stop codons were sub-optimal. Mike offers an explanation of why they aren't, and it ties in with his thoughts on C-T deamination.

    Bilbo:
    "True, but I was just trying to understand the data at that time. Let me offer an ID perspective, now. A knowledgeable designer, realizing that there was an affinity between some of the amino acids and their codons or anti-codons, designed the tRNA synthases so that the amino acids would bind to tRNAs with those codons, thus avoiding potential problems of interference by binding to the wrong tRNAs."

    That's just an unsupported and untestable assertion. I can as easily and as justifiably claim that a knowledgeable designer would have nothing to do with any of this, and thus that life as we know cannot be designed.

    I'm working with the hypothesis that a knowledgeable designer did choose to design our kind of life, and tried to come up with an explanation as to why she would choose to have the tRNA synthetases bind those amino acids that showed affinity with certain tRNAs to those tRNAs. Is is testable? Can we change the synthetases in vitro and see what happens?

    My main problem with the DM (as I'm reading it here) is that it seems to be a needless multiplier. If one is going to see an analogy, then rationality and foresight, at the very least, are going to be foregone conclusions. The same will usually be true for discontinuity (a property that is folded integrally into most analogies in the first place).

    I'm not sure I agree. Let's say on closer inspection the face on Mars still looks designed. We land and explore, and find no other designed-looking objects. We have high scores for Analogy and Discontinuity, but low for Rationality and Foresight. I don't think our conclusion of design will be as strong, and doubters will be tolerated.

    Seeing as Mike Gene thinks I accused him of child abuse and poor parenting...I sort of suspect that people are confusing me with someone else.

    Did you?

    ReplyDelete
  30. Hi Bilbo,
    Returning to this discussion:

    You said: ” I found another library. Data for discontinuity: Apparently 12 amino acids show no affinity for the (anti)codons. No remnants of an earlier code found. At best, highly speculative scenarios of how the code could have evolved from an RNA world.”

    I don’t know where you got this from, but its wrong. The facts are that 11/12 of the amino acids that were tested showed a decided and significant affinity for RNAs that possess codon and/or anticodon. The other amino acids have not been tested – not for philosophical or theological reasons, but because the method cannot be used with these amino acids. I discuss this in more detail at http://fluorescentflicker.wordpress.com/2007/12/30/bradfords-hammer/#comments .

    Quoting Mike Gene:
    "When it comes to Discontinuity, we must pay tribute to the fact that there are dozens, perhaps hundreds, of papers where scientists have speculated about the evolutionary origin of the code and gathered circumstancial evidence to support their respective cases. Neverthelss, the various explanations remain largely in the realm of speculation....the code is universal, lacking any trace of simpler precursors or permutations spawned from such ancient precursor states....if the code was indeed spawned from non-teleological forces acting over great spans of time, we would predict the existence of such permutations someplace in the biological world. That they do not exist argues against such an origin. As a result of the above considerations, I give the code a Discontinuity score of 2."

    If it hadn't been for the hundreds of papers by scientists speculating on a non-teleological origin, Mike probably would have given it a higher score.”

    This is pretty lame. Basically, Mike Gene is saying, because the scientific community hasn’t come to a clear and unequivocal model for each and every step in the OOL, then the positive results of Yarus and coworkers may be ignored. As I have been saying in this discussion here, this fairly screams “don’t try to confuse the issue with facts”. This is, in a nutshell, the DM. An attempt to substitute vague and unfounded suspicion for hard experimental data.

    You said: ”I think I've explained why I wouldn't. But I understand why you do, and I don't think you are entirely unjustified. I do suggest a -1, instead of a -2.”

    Thanks. This would bring us to an examination of the scale of the DM. But I have better things to do with my blogging time than quibble thusly. So I’m happy to leave things at this significant point of agreement. (It’s significant in that we can see what the other is saying, and comment about the specific issue.)

    Incidentally, as most of us have found, the ARN search tool is really bad. I’ll keep trying to find the specific reference I alluded to before, but will mention as an aside that I have found more recent studies that also fill in the “gap” between Yarus and the genetic code as we know it. Maybe someday when I get caught up on my blog, I’ll put something together on this.

    “I've only read Franklin Harold's book, The Way of the Cell, and I doubt he was familiar with Yarus's work. It could be that Harold would have changed his mind. There's also Maynard Smith and Stuart Kaufmann. Again, they might not have known about Yarus, but Kaufmann makes the point that the arbitrariness is apparent from the fact that the amino acids bind to the stem of the tRNAs, and not the anticodons.”

    Obviously, Kaufmann was speaking without the knowledge of Yarus et al.

    “Art said...
    Mike Gene's blog entry has nothing to do with the matter I mentioned.

    Well, if I recall, you thought the 3 stop codons were sub-optimal. Mike offers an explanation of why they aren't, and it ties in with his thoughts on C-T deamination.”

    My point is that, from a purely conceptual and hypothetical point of view, one stop codon works as well as three. My point has nothing to do with the base content of the stop codon, and is as applicable to the stop codon as UAA as CCC (or whatever…).


    “I'm working with the hypothesis that a knowledgeable designer did choose to design our kind of life, and tried to come up with an explanation as to why she would choose to have the tRNA synthetases bind those amino acids that showed affinity with certain tRNAs to those tRNAs. Is is testable? Can we change the synthetases in vitro and see what happens?”

    Yes. But what this has to do with your hypothesis is a bit unclear.

    “I'm not sure I agree. Let's say on closer inspection the face on Mars still looks designed. We land and explore, and find no other designed-looking objects. We have high scores for Analogy and Discontinuity, but low for Rationality and Foresight. I don't think our conclusion of design will be as strong, and doubters will be tolerated.”

    I think I see your point. I remain skeptical that a similar scenario has, or will, play out when it comes to life as we know it.

    “”Seeing as Mike Gene thinks I accused him of child abuse and poor parenting...I sort of suspect that people are confusing me with someone else.”

    Did you?”

    No.

    ReplyDelete
  31. Yeah, I miss Matty. So does Arp of Jockstrappe.

    ReplyDelete
  32. 11 out of 12 amino acids? I guess I counted wrong at iscid (not surprising, if you know me). That's much more impressive and I would definitely give you a -2 on Discontinuity.

    However, the question remains, could this be explained from a design perspective? Would a knowledgeable designer (by hypothesis), realizing that the amino acids have an affinity for specific codons have a good design reason for designing the appropriate synthetases, that would bind those amino acids to the tRNAs having those codons? For example, if the designer used synthetases that bound those amino acids to different tRNAs, would the affinity with the other codons interfere with the binding between the synthetases and the tRNAs? That doesn't sound clear. Let's use a single example: Say amino acid x has an affinity for the anticodon of tRNA-x. However, its synthetase binds it to tRNA-y. Would amino acid x's affinity to tRNA-x interfere with its binding to tRNA-y? If so, then a knowledgeable designer would have designed it so the synthetase would bind amino acid x to tRNA-x, instead of to tRNA-y.

    So the experiment I propose would be to change the synthetase so that an amino acid binds to a different tRNA, one that it doesn't have an affinity with; put it in with it's old tRNA and its new tRNA, and see what happens. Would we notice a difference in binding time or tendencies? Would it bind to the new tRNA, only slower? If so, then we would have a design explanation for why the 11 of 12 amino acids bind to tRNAs that have anticodons with which they have affinity.

    Would this disprove the non-teleological explanation? No. We would have ambiguous results, unsurprisingly. Which is why the "facts" don't always give us the clear conclusions we think they do. And that's why Mike Gene's use of the duck/rabbit picture.

    Yeah, on re-reading the article on C-T deamination and the stop codons, I'm not sure it explains the 3 codons. However, I'm curious what the author was getting at about most of the codons in prokaryotes being a certain type.

    ReplyDelete
  33. Does Matty ever show up at ARN? I haven't been there for a while.

    ReplyDelete
  34. I'll be gone for a few days, Art. After finally understanding that Yarus found that 11 out of 12 amino acids have affinity for their (anti)codons, I understand your frustration with Mike Gene for not talking about more specifically in his book. However, given that he scored pseudogenes as a -4, PCP pathway as -2, the vertebrate eye as 0, and the genetic code a whopping (notice the sarcasm) +3, I don't think we're dealing with a guy who ignores the scientific evidence. I think he deserves better from you.

    I'll be gone until next week.

    ReplyDelete
  35. Hi Bilbo,

    I'll be gone for a few days as well, during which time this thread will possibly wander off the front page, and thus into history (or perhaps oblivion). If you wish to keep this back and forth going, you should start a discussion on your own blog (TT doesn't work as I've long ago been disinvited from there).

    I hope you will take note of the inconsistency on Mike Gene's part when it comes to Yarus et al. He is willing to look at data that pertains to some issues. But not in this case. IMO, this betrays a bias that takes precedence over other arguments, a prejudice grounded in the proposition that the origins of life cannot ever be allowed to come into the field of scientific play. The stereochemical basis for the genetic code is threatening to our telic thinkers because it is a chink in their "OOL NO NO NEVER!!" facade, a "fortress" in which ID seems more and more entrenched. They will not be able to think clearly and objectively about research like this - it crosses a line that must not be crossed.

    Whatever the reason for your hiatus, best of luck.

    ReplyDelete
  36. Hi Art,

    I didn't know you had been "disinvited" from TT. That's problematic for me, since I consider you one of the best critics around. Without good critics, what's the point of discussing ID? And what's the fun of it?

    You write,
    I hope you will take note of the inconsistency on Mike Gene's part when it comes to Yarus et al. He is willing to look at data that pertains to some issues. But not in this case.

    Ummm, let's see. He probably lowered his Discontinuity score because of it. Not low enough for you, but then he's working from a Rabbit perspective, and you're working from a Duck perspective, so I think that explains it. And did Yarus's work also cover amino acids' affinity with all their tRNAs? Because Mike wrote:

    "...there is no law of nature which determines that the codon UCU must represent the amino acid serine, because other codons can also represent this amino acid, for example, AGC." (p.71 of the DM

    ReplyDelete