|
Author
|
Topic: Cytosine Deamination from Both Sides
|
Pim van Meurs
Member
Member # 541
|
posted 25. April 2003 02:31
Mike: It is also often claimed that ID cannot generate any scientific research (i.e., ID is a "goddidit" explanation and nothing more). Would you agree that I have turned back that criticism?
I do not think that you have. By pursuing what Dembski refers to as a regularity to explain the observations it seems hard to infer design.
I would also not refer to ID as a 'goddidit' explanation but more as an eliminative argument which seems run the risk in many instances to become an 'argument from ignorance' or 'god of the gaps' argument.
quote:
To attribute an event to design is to say that it cannot be reasonably referred to either regularity or chance.
...
The concept of design that emerges from the design inference is therefore eliminative, assertion of an event what it is not, not what it is.
W. Dembski
As Ratzsch argues "[That] Dembsksi is not employing the robust, standard, agent-derived conception of design that most of his supporters and many of his critics have assumed [seems clear]"
I would say that your approach has little to do with ID (as proposed by Dembski, Johnson, Behe, Wells, Meyers etc) or the ability of ID to propose original research. Your proposal at most uses design as a (personal) tool for understanding the implications for a natural pathway and proposing fully natural pathways. In other words, ID is totally extraneous to the issue.
Original ID research in my opinion would include a proposal for an ID pathway for instance. Front loading merely moves the problem of ID back into the past without resolving any of the relevant ID issues. On the other hand, appeal to interventions opens up a whole other can of worms...
Note that I am not saying that using analogies to design is useless, it has helped over time people formulate many hypotheses "God does not throw dice", "Planets move in 'perfect' circular orbits", and countless more. What I am wondering how such 'analogies' can be extended to include ID?
[ 25. April 2003, 02:54: Message edited by: Pim van Meurs ]
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 25. April 2003 02:41
I'm afraid I haven't read the peer-reviewed literature that suggested that cytosine wouldn't have been chosen by an engineer. If you are trying to refute that statement only, I agree that you have succeeded: we are now in doubt over whether an engineer would have chosen cytosine or not, since C->T transitions can do interesting-looking things. This doesn't mean that an engineer would have chosen cytosine--just that we don't have enough information to tell.
I suppose this is a case where the motivation was specific to ID, but given that the only conclusion of the research is to retreat from what I imagine was an off-hand comment in a discussion section, I'm afraid I can't be very impressed. It is better than nothing, and personally, I think it's neat, but as it stands it's more of a scientific curiosity than anything else.
Then again, I'm not very enthusiastic about the relevance of most origin-of-life type research, since it is typically extremely speculative.
Added in edit: Well, it looks like there was an opinion piece on the problems of cytosine deamination in a review journal, not an off-hand comment in a discussion section. I am still not impressed, but I might be slightly more so if there were some indication why having a 100-1000fold excess of removed helix breakers and stop codons was advantageous.
[ 25. April 2003, 03:22: Message edited by: Rex Kerr ]
IP: Logged
|
|
GP
Member
Member # 570
|
posted 25. April 2003 09:54
I am a little short on time today, so let me put out three observations:
1) The IHE hypothesis posits that the standard genetic code was intentional. What is observed is essentially the residues ordered by some criteria (in this case "hydrophobicity," though I believe that there isn't a standard hydrophobic scale) and then matched to a key with a codon of 3 bases. What would be interesting to me is a demonstration that any random ordering of the keys would also result in at least one of the bases correlating with some criteria on corresponding amino acid. My guess is that because of the degeneracy of the code, this likelihood is increased. Consequently, it may have been just as likely to see a code developed so that X->Y transitions resulted in an increase in hydrophobicity. I am not fond of the frozen accident hypothesis, but at the point I am not obligated to rule it out even by the IHE hypothesis.
2) The latest posts suggesting that 2 or more rounds of deamination can achieve all hydrophobic residues is confusing, especially since the codes themselves represent a connected network (i.e. a graph where two codons are connected if they share two base pairs). In a two-coloring scheme (i.e. essentially "hydrophobic" and "nonhyrdophobic"), I don't see the point of showing how one can get from one type to the other in a couple of jumps on this network [more on this point below].
3) Given N cytosines in a sequence, let us say that the engineer was hoping for a specific set of Cs to become Ts. It seems to me that the cost of achieving a desired state increases exponentionally with respect to ratio of the number of possible states and the number of desired states. For N cytosines, I am guessing that there are 2exp(N) possible C->T states. Say the engineer was only expecting 1 out of those states to occur. That suggests that he is constrained by the number of cytosines to use (i.e. how big N can be), before the odds of some other mutational event prevents the desired state from ever "unpacking" becomes much greater than the odds of that one specific C->T transition from happening. Furthermore, if multiple rounds of deamination are required in a codon (e.g. CCx codons), the odds seem to me to drop further, especially if the intermediate residue is polar (or are you also saying that this intermediate is part of the engineer's Plan?). Even then, this does not appear to me to be an efficient manner of design.
PS: Just one more point: quote:
But this objection is built upon many assumptions. Remember, I am trying to keep such assumptions to a minimum and thus consider only a human-like intelligence behind the designs.
Mike, you said that you assumed a "human-like intelligence" is minimal. How so?
[ 25. April 2003, 10:26: Message edited by: GP ]
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 25. April 2003 15:39
Rex,
As you noticed, the original argument about an engineer not employing cytosine was not an off-hand comment, but was part of an argument from an opinion piece in Nature Reviews . The argument was mentioned in the abstract and expanded on in "Box 1" of the article. In fact, in that argument, the authors conclude that frozen accidents "are therefore a consequence of evolution through tinkering."
This was not a simple case of having a motivation specific to ID. Motivation alone will not get you anywhere. First, there is a different point of emphasis involved. That is, since all evidence is interpreted data, here was a case where data (cytosine deamination) was interpreted differently. Among non-teleologists, there is a tendency to terminate their investigation at the level of the frozen accident. In fact, as the authors argue, this is essentially a prediction of non-teleological evolution. A teleologist, in contrast, will have a tendency to take a closer look to see if any patterns might exist around the "frozen accident." To take this closer look, one has to think like an engineer and have some type of explanatory template as a guide. In the case of thinking like an engineer, I am not one. However, it would seem me that a good engineer is one who can turn a problem into an asset. After all, good things also come with cytosine (the parity code). And it would seem to me that a biotic engineer, realizing the inevitability of evolution, would look for ways to make use of evolution. Perhaps engineers on this forum can address this point.
As for the template, the concept of front-loading evolution had already been gelling in my mind and it should be clear from my first essay how the concept of front-loading led me to consider the genetic code might be exploiting cytosine deamination (see figure 7 from BR. No18).
As for these dynamics being mostly a scientific curiosity, I fully agree. I have never tried to make more of this than what exists. As I explicitly stated, "Nevertheless, it looks as if I once again need to remind people that my stuff is not only tentative, but represents the very first steps of an investigation." Any investigation begins with curiosities that catch our attention. The key is follow-up. But it is worth noting that there are many critics of ID who believe that ID is incapable of even identifying curiosities about the processes of life.
Nevertheless, it would seem to me that since the peer review process allowed the argument about engineers, evolution, and cytosine deamination into the pages of a well respected scientific journal, don't you think the same literature should allow the description of this curiosity? After all, science proceeds once the idea is implanted into the larger community. Several people keep challenging ID proponents to put their arguments in the mainstream literature. I'm just trying to see what the same people say about these essays. Does it deserve a wider hearing? Or should it be excluded from such a hearing?
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 25. April 2003 15:49
GP: Mike, you said that you assumed a "human-like intelligence" is minimal. How so?
The thesis is a modest extension on Crick and Orgel's hypothesis of directed panspermy. Thus, I can easily envision a human-like intelligence seeding planets with designed life-forms (something we ourselves may end up doing in the next century or so). However, subsequent direct interventions would suggest that our intelligent agents have overcome the problems posed by deep time. Thus, I try to envision how one could design something such that it influences the future, rather than simply assuming the designers continually intervene over time. Front-loading is about using the present to impose direction on the future. It entails the use of regularities for these objectives.
Today, the United States government is trying to front-load the Middle East by implanting a democracy in Iraq. That is, instead of choosing a strategy where the United States will directly control and manipulate Middle Eastern countries for the next few centuries, it is trying to rig the political conditions such that intervention is not required. A certain amount of mess is to be tolerated as long as the objectives are approximated. Why? Because in a long-term geo-political (and economic) sense, it is not really possible to for the United States to directly run the Middle East. Front-loading becomes the only option when direct, sustained manipulation is not possible. And likewise with our biotic front-loaders, it may simply not be possible to exert direct control over everything in the biosphere over billions of years.
Of course, if the designer is God, things change. But even here, it now becomes a question of theology and whether God is the type of Being who would play puppet-master with the creation. In other words, the notion that a divine designer would exert direct control (in a way that yersinia suggests) over every novel biotic change rather than use front-loading is anything but well-established.
Does any of this help?
I'll try to get to your other points as soon as possible.
[ 25. April 2003, 16:02: Message edited by: Mike Gene ]
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 25. April 2003 16:08
Pim,
There is not one, and only one, way to think about design. I simply take the hypothesis that the original life forms were designed and begin thinking about the implications of such a possibility. It clearly leads to questions, hypotheses, and thus research (I have provided many illustrations with many more to come). If you don't agree, we'll just have to agree to disagree.
But you do write," Original ID research in my opinion would include a proposal for an ID pathway for instance."
What do you mean by an "ID pathway?" If you rule out the employment of regularities, are you suggesting I need to mimic Indiana Jones and go on a quest for the lost protocols for life's design? After all, isn't an 'ID pathway' otherwise known as a blueprint, procedure, protocol, and/or recipe?
In an experiment (experiments are the product of design), the "ID pathway" is found in the materials and methods. Can you recreate the materials and methods purely from the results? And what if you have no solid experience ever doing science? Has the loss of the materials and methods section become even more problematic?
[ 25. April 2003, 16:13: Message edited by: Mike Gene ]
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 25. April 2003 18:21
Mike,
Opinion pieces are not usually peer-reviewed. Typically the editors will check to make sure nothing too outrageous is said; but the review is only to check that it is within the bounds of reasonable opinion. (Reasonable people can have different opinions.)
Also, the opinion/review piece made a number of speculative but biologically relevant points. First, it reviewed the evidence that deamination of cytosine gave uracil, which meant that mutations couldn't be fixed. This is obviously detrimental to fitness, and hence one would expect that organisms that replaced uracil or cytosine would have a selective advantage; apparently uracil was replaced, which at partially solved the problem (in the context of error-repair mechanisms).
Unfortunately, your work doesn't make a strong enough link to biological relevance. Specifically: could it possibly be advantageous to have a ridiculously high C->T substitution rate? Is there any reason to believe that C->T will be particularly useful to front loading as compared with any other system? Why build very high levels of mutation into the system instead of adding genes to induce mutation?
What you have now might be suitable for part of an opinion piece (or speculation in a discussion secton), but it's rather lacking in biological relevance. For publication in a biological journal, that's the gold standard.
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 26. April 2003 14:49
GP,
You raise some throught-provoking points.
The IHE hypothesis is little more than a scientific curiosity. As such, it clearly needs more consideration. Your suggestion about a demonstration that showed any random ordering of the keys would also result in at least one of the bases correlating with some criteria on corresponding amino acid is a good test and again shows how ID can lead to research projects. But therein lies my dilemma. I neither have the time nor sufficient skills for this type of analysis. Normally, one should be able to tap into the global scientific mind and perhaps recruit those with the sufficient skills and similar interests, perhaps even putting a grad student or post-doc onto the scent. Unfortunately, given the sociological dynamics centering around the issue of ID, the inability to take this next step is a very real problem.
As Rex explains, the hypothesis does not even deserve a wider hearing via the scientific literature.
There are other considerations, however, beyond such a test. First, keep in mind that several studies have already indicated the genetic code is highly optimized to resist deleterious mutations. Thus, it becomes even more interesting that such a pattern of radical changes are embedded within a code that is rather optimal overall. Thus, our random ordering of keys would have to pull out both features. Secondly, the IHE pattern is also coupled to a form of base substitution that is, in the least, among the most common. Our random ordering of keys should also mimic this feature. Thirdly, I'm not quite sure the comparative test you mention would be all that definitive. Recall that my original thesis for the utility of cytosine deamination is not a generic "stand alone" but is tied to the originally specified conditions. And what appears to be the insurmountable problem is a precise definition/description of such a historical state. Put simply, when held up against the constitution of the original states, the effects of C-T transitions, may have been wildly uncanny.
So in the meantime, there is an interesting historical development of this thesis on my part, which explains why it probably resonates for me personally. Remember the original impetus. I had pointed to a paper which argues that the bases of DNA work like a parity code, thus pulling out the rug from underneath the frozen accident thesis for the existence of A,G,C/U,T among the genetic material. A paper was then cited arguing that no engineer would have included cytosine in the DNA. Taking the thesis that a design problem presents design opportunities, and evolution is inevitable among replicating entities, I took a closer look at the effects of cytosine. I immediately steered to the genetic code, as I have already argued that it was intentional on other grounds. And low and behold, when a very simple analysis was done, a clear pattern emerged. C-containing codons coded for amino acids that seemed to be randomly distributed in terms of their hydrophobicity. And C-T transitions converted them into a tightly clustered group all at the extreme of the hydrophobic scale (see Figure 4 from BR No. 18). What's more, there seems to be a very solid pattern toward moving to residues more predisposed to forming secondary structures.
It could, of course, be a coincidence that everything fell into place like this.
That suggests that he is constrained by the number of cytosines to use (i.e. how big N can be), before the odds of some other mutational event prevents the desired state from ever "unpacking" becomes much greater than the odds of that one specific C->T transition from happening.
Front-loading would not depend on achieving a highly specified state in a specific place and time. If the secondary design is unpacked in one genome at one point in time before it is advantageous, it still remains buried in millions of other genomes spread about the globe. It remains in reserve unless it is wiped from the face of the planet. You can keep it from being wiped away by burying it in gene whose original function/sequence is largely conserved.
As for the efficiency of designing evolution, it is extremely difficult to make any judgment in these regards given that none of us really have any experience designing organisms to evolve over deep time. It would seem, however, that compared to a hypothetical state where all nucleotides underwent substitutions equivalently, and where amino acid substitutions, under the guidance of the genetic code, appeared random, the existing state would be more efficient if one was attempting to rig evolution. Keep in mind, however, that I think that the cytidine deamination/IHE story is only a small part of the story. I liken it to finding tata boxes - an important part of transcription initiation, but far from being the whole story. Thus, rather than throw in the towel because it seems inefficient and could all be a coincidence, I'll keep digging to the best of my limited abilities. ![[Wink]](wink.gif)
IP: Logged
|
|
GP
Member
Member # 570
|
posted 26. April 2003 18:04
Hello Mike,
Thank you for your replies. Allow me to reply to your posts out of order and see if I can't construct a more coherent narrative. You wrote: quote:
The IHE hypothesis is little more than a scientific curiosity. As such, it clearly needs more consideration. Your suggestion about a demonstration that showed any random ordering of the keys would also result in at least one of the bases correlating with some criteria on corresponding amino acid is a good test and again shows how ID can lead to research projects. But therein lies my dilemma. I neither have the time nor sufficient skills for this type of analysis. Normally, one should be able to tap into the global scientific mind and perhaps recruit those with the sufficient skills and similar interests, perhaps even putting a grad student or post-doc onto the scent. Unfortunately, given the sociological dynamics centering around the issue of ID, the inability to take this next step is a very real problem. As Rex explains, the hypothesis does not even deserve a wider hearing via the scientific literature.
I will not speak for Rex, Mike. However, judging from your political analogies and the posts to me, I sense that you are keenly aware of the sociological dimensions with respect to ID. I am also, however, bound by the rules set by the Moderator of this forum that forbid me to discuss such matters directly. So, let me address your concerns via science. You write specifically that the IHE hypothesis is little more than a scientific curiosity. To this statement, I largely agree. But the interesting observation that specific bases in a codon correlate with a specific amino acid property by itself cannot bear the burden of your extrascientific concerns. The question I would ask myself here is whether or not your teleological perspective was required to make this observation. Offhand, I see no reason why it should. In fact, observations about the genetic code similar to your IHE hypothesis have existed for quite a while, and if I am not mistaken researchers are quite aware that hydrophobicity is a preserved property in the SGC. The properties and the evolution of the SGC remain fertile areas of research. In fact, later in your posts, you pointed out that studies were done with respect to the optimality of the SGC, and I see no evidence in that case either that one must consult a teleological perspective to conduct such a study. Consequently, I fail to identify with your concern that the global scientific mind cannot be tapped to conduct the research you are interested in.
Before I continue, let me say that I can sympathesize with your dilemma in which you promote a research topic where you lack the skills and the time to pursue effectively. I do not believe you are alone in this regard, and this problem cannot be solved by voluntarily isolating oneself from the scientific community, for whatever ideological reason. (I assume, of course, that you haven't tried publishing these results yet.) However, if you truly think there is no effective way of selling your ideas so that the scientific community may find it worthwhile to pursue them, then I submit that you simply let others do it in your place.
Let me then qualify the elements of the IHE hypothesis that I do not find a worthwhile scientific pursuit. Earlier I had asked you in what respect your assumptions about the engineer behind your front-loading thesis were minimal. You wrote: quote:
The thesis is a modest extension on Crick and Orgel's hypothesis of directed panspermy. Thus, I can easily envision a human-like intelligence seeding planets with designed life-forms (something we ourselves may end up doing in the next century or so). However, subsequent direct interventions would suggest that our intelligent agents have overcome the problems posed by deep time. Thus, I try to envision how one could design something such that it influences the future, rather than simply assuming the designers continually intervene over time. Front-loading is about using the present to impose direction on the future. It entails the use of regularities for these objectives.
First of all, a brief comment about directed panspermia. I am only vaguely familiar with the hypothesis, but I am not aware that it is fertile area of research. Are there any recent peer-reviewed articles that you would like to point me to, so that I may catch up? But, I guess I need not point out to you (with all of your concerns about tapping into the global scientific resources) that a modest extension of a controversial thesis inherits all of the criticisms of that thesis. In any case, I agree with you on the prospective utility of the front-loading concept, but it seems to me that the problems lie with the retrospective utility of discovering the front-loading effects. In other words, can we in fact reconstruct an initial state, or is it always an assumed entity? By your own words, I think most of the thesis derives from the latter approach, when you say: "And what appears to be the insurmountable problem is a precise definition/description of such a historical state." Correct me if I am mistaken, but it seems like the approach you have taken is to assume an initial state, and then match your observations with regards to possible evolutions of that state.
I find this approach a riskier proposition when you combine an already putative initial state with derived intentions of an agency (i.e. our engineer). You write: quote:
Today, the United States government is trying to front-load the Middle East by implanting a democracy in Iraq. That is, instead of choosing a strategy where the United States will directly control and manipulate Middle Eastern countries for the next few centuries, it is trying to rig the political conditions such that intervention is not required. A certain amount of mess is to be tolerated as long as the objectives are approximated. Why? Because in a long-term geo-political (and economic) sense, it is not really possible to for the United States to directly run the Middle East. Front-loading becomes the only option when direct, sustained manipulation is not possible. And likewise with our biotic front-loaders, it may simply not be possible to exert direct control over everything in the biosphere over billions of years.
Of course, if the designer is God, things change. But even here, it now becomes a question of theology and whether God is the type of Being who would play puppet-master with the creation. In other words, the notion that a divine designer would exert direct control (in a way that yersinia suggests) over every novel biotic change rather than use front-loading is anything but well-established.
Before I go on, let's remind ourselves that this was written in response to a question about the minimalist nature of your human-like intelligence. I think only the second paragraph really addresses what I was looking for, but even then I am not sure if it does the job. Is God an example of a human-like intelligence that is minimal? How so? I guess I am confused about what essential properties are minimally required to qualify as a human-like intelligence. The reason I would demand such a characterization is that the model of your engineer seems inextricably tied to the predictions that you are making. After all, you are saying to me: "I try to envision how one could design something such that it influences the future, rather than simply assuming the designers continually intervene over time." In other words, the IHE hypothesis makes sense only in light of an engineer that designed for the purpose of minimizing inteventions of his inventions across time. Or put more pointedly, you assume the engineer is not a micromanager.
So, I guess the question here is whether or not that characteristic is part of the assumption of what constitutes a human-like intelligence, or if it is derived from independent considerations. Personally, I think it is consistent with the approach I described above, in which you assume an initial state (i.e. front-loaded organism plus engineer), and then interpret the observations accordingly. It is also consistent with the political analogy that you went briefly into. Here, we assume that "the United States government is trying to front-load the Middle East by implanting a democracy in Iraq" can prospectively lead to an observation (perhaps realized sometime in the future) that the Middle East has a stable political situtation. This conjecturing is all sensible, but I must ask if it makes good science. Let's remind ourselves that the biotic problem at hand is not at all the one that you suggested in the analogy. If I may borrow the analogy, I think the question we are trying to ascertain is something akin to: Is the current American foreign policy an intentional result of Ghengis Khan's imperialism centuries ago? Or was this Gulf War inevitable (i.e. front-loaded) from the first Gulf-War, as the political pundits claim, or was it front-loaded by the Crusades, as some Arabs claim? I apologize if this is not where you had intended this analogy to go. It is difficult to see how even in our own time how these kinds of questions can be reliably ascertained. However, answers can always be generated depending on the model of the front-loaded system. For instance, you later write that it is up to theology to determine whether or not God is a micromanager. Suppose in fact that theology determines God to be a micromanager, intervening across deep time. Does the validity of your front-loading thesis actually depend on reconciliation with theology or vice-versa? In any case, I hope I have made more clear my initial question about your human-like intelligence. In particular, I wanted to know if you actually try to minimize your dependence on the model of the engineer to interpret the results. At present, I do not see this effort made.
My remaining points to you were specifically about whether or not an engineer in fact would have designed with cytosine deamination as a mode of unpacking hidden states. Of course, I was initially thinking that I was a member of this class of human-like intelligence, but now I am not so sure if you had me in mind. As you claim not to be an engineer, maybe this whole exercise of figuring out whether or not a particular design strategy is sound becomes a little more difficult. But, let's take a quick look at what you had to say: quote:
Front-loading would not depend on achieving a highly specified state in a specific place and time. If the secondary design is unpacked in one genome at one point in time before it is advantageous, it still remains buried in millions of other genomes spread about the globe. It remains in reserve unless it is wiped from the face of the planet. You can keep it from being wiped away by burying it in gene whose original function/sequence is largely conserved.
OK. Then it seems that this engineer really did not have any specific hidden states in mind with respect to cytosine usage. After all, if the secondary design need not have a selectable function until much later in time, it makes little sense why the engineer didn't directly implement the design in the first place. Otherwise, if at least one of the selectable function was a necessary intermediate, then he faces a daunting exponential problem of ensuring its arrival. Furthermore, as you mentioned, C->T are not the only mutational events. Are we also considering the possibility of deletions? recombinations? How can this engineer account for all those possibilities while waiting for the states to unpack? Perhaps I am supposed to read into this statement of yours: quote:
As for the efficiency of designing evolution, it is extremely difficult to make any judgment in these regards given that none of us really have any experience designing organisms to evolve over deep time. It would seem, however, that compared to a hypothetical state where all nucleotides underwent substitutions equivalently, and where amino acid substitutions, under the guidance of the genetic code, appeared random, the existing state would be more efficient if one was attempting to rig evolution. Keep in mind, however, that I think that the cytidine deamination/IHE story is only a small part of the story. I liken it to finding tata boxes - an important part of transcription initiation, but far from being the whole story. Thus, rather than throw in the towel because it seems inefficient and could all be a coincidence, I'll keep digging to the best of my limited abilities.
Positing an intelligent catalyst is always possible. From what I have seen you write here, I have no reason to believe that there are in fact any constraints on the causal efficiencies of directly implementing a feature of biotic reality. But the risk that I believe you run is an over dependence on a model of the intervening agency. I have illustrated this problem in the paragraphs above. Moreover, the narrative that you build becomes frustratingly dependent on building a complete picture of this entity's intentions, one which a non-teleologist is free from considering directly. This problem manifests especially when you ask that I suspend judgment on what I perceive as a lack of efficiency but in the same breath suggest that your perceived efficiencies are more believable than current models.
I won't be able to respond to any further comments that you may have for several weeks, but it was a pleasure.
GP
PS: Since this will be my last post in a while, I wanted to add one additional observation, before I sign off. You wrote: quote:
What's more, there seems to be a very solid pattern toward moving to residues more predisposed to forming secondary structures.
I'd like to know how you are so certain of this bias, given that the field is still struggling to build models and identify the determinants of secondary structures (and more importantly, higher order functional structures).
[ 26. April 2003, 19:41: Message edited by: GP ]
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 27. April 2003 16:04
Rex,
It is interesting that you point out editors " will check to make sure nothing too outrageous is said" and also ensure that the review/opinion piece " is within the bounds of reasonable opinion." It would seem obvious to me that any ID speculation would run into trouble on this basis alone.
You write: Unfortunately, your work doesn't make a strong enough link to biological relevance. Specifically: could it possibly be advantageous to have a ridiculously high C->T substitution rate? Is there any reason to believe that C->T will be particularly useful to front loading as compared with any other system? Why build very high levels of mutation into the system instead of adding genes to induce mutation?
Unfortunately, the concept of biological relevance has a distinct subjective element to it, where different people arrive at different judgments of relevancy. In this case, there is evidence that cytosine deamination can influence genomic organization (two examples cited above) and is also more common as a function of transcription. Furthermore, I argue the biological relevance of the IHE here.
As for your questions, all hypotheses usually raise more questions than answers. All that you have shown is that this hypothesis opens many avenues of research. Let's look at them:
1. Could it possibly be advantageous to have a ridiculously high C->T substitution rate?
That would depend on the sequence, function, and environmental conditions in which the mutation occurred. It might also involve targeting such mutations. For example, there is growing evidence that connects transcription and deamination in somatic hypermutation involved in immunoglobulin class switching. And Shapiro has already published speculations in the literature using this example as a model for natural genetic engineering. Furthermore, in the link provided above, there is a discussion of no less than 441 C to U changes in ORFs of Arabidopsis mitochondria. I speculated further above: "I provided an example whereby RNA editing exploits cytosine deamination to demonstrate the reality of the IHE. It's a proof-of-principle demonstration that adds plausibility to my hypothesis. The advantage to RNA editing, rather than mutating the genome, takes us to the soft-wiring ability of RNA (to be discussed later) and the fact that RNA editing often allows a cell to expresses two variants of a gene at the same time. In such cases, no single example of a C-T transition at the genomic level may be beneficial. But through massive editing, multiple C-T events in multiple genes might impart significant protein re-design such that a benefit it imparted. In fact, in a quasi-IC likeness, the beneficial state might depend on multiple simultaneous mutations in the context where expressed original sequence is likewise maintained."
2. Is there any reason to believe that C->T will be particularly useful to front loading as compared with any other system?
Not yet. However, since nobody else is thinking in terms of front-loading evolution, there is no other system to compare it to. Thus far, it appears to have more potential than a situation where all nucleotide are substituted equally and their consequent amino acid changes also appear random with regard to amino acid properties. The point remains is that the C-T transitions remains a fertile ground for inquiry from an ID perspective.
3. Why build very high levels of mutation into the system instead of adding genes to induce mutation?
Why build a system by adding genes to induce mutations rather than one with an intrinsic and rather specified high level of mutations? Building high levels of mutations into the system creates a flux that can be regulated by error-correcting genes, that in turn, could be subject to regulation. On its surface, the fact that the mutational effect is buried deeply into the wiring of life suggests that front-loading may reach further out that I anticipate. One thing a designer can count on is this - DNA, the genetic code, and amino acid properties will persist in a rather universal way as long as life persists.
The bottom line is that this is a question to be investigated, not a reason to keep a hypothesis out of the literature that is read by investigators. It appears to me you expect my hypothesis to enter the fray as a extensively tested and supported theory.
What you have now might be suitable for part of an opinion piece (or speculation in a discussion secton), but it's rather lacking in biological relevance. For publication in a biological journal, that's the gold standard.
And therein lies the problem for ID and the scientific literature. In the end, subjective judgments concerning whether something is within the bounds of "reasonable" opinion and whether something is "relevant" to the journal and its mission will come into play. And those reasons are sufficient to keep a extremely modest ID proposal out of the literature.
Anyway, if relevance remains the issue, perhaps we can consider other published speculations. For example, consider the paper where it is argued that iron sulphide honeycombs played a role in the origin of life. I see no demonstration that such structures are biological relevant. In fact, Thomas Cavalier-Smith described the hypothesis as "impossible." And rather than being overly concerned about a justification of relevance, we read this:
quote:
"I think it's a beautiful thing - it's important to have all-embracing theories," says evolutionary biologist Ford Doolittle of Dalhousie University, Halifax, Canada. We'll never have much definite information on the origin of life, he says. "But then, just because we'll never know why the Roman Empire fell doesn't mean it isn't worth talking about."
and this
quote:
It may be that no theory is going to fit all the evidence. The trick is to pick which bits to ignore, says John Raven of the University of Dundee, UK. "To create a coherent hypothesis we have to say 'this bit of data doesn't fit, but we're going ahead anyway'."
Perhaps what I need from Rex is a rather well-defined description of biological relevancy and what factors an explanation must minimally possess to be judged relevant enough to get a mere hearing among the scientific community.
[ 27. April 2003, 16:08: Message edited by: Mike Gene ]
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 27. April 2003 21:48
GP,
I never claimed that my teleological perspective was "required to make this observation." It's just a simple fact that my teleological perspective was behind making this observation. The genetic code, amino acid properties, and cytosine deamination have long been well-understood. Yet as far as I can tell, no one else has ever made the connection that I made. I do, however, appreciate your suggestion about consulting others since teleology is not required. But from my perspective, given that a teleological perspective was extremely useful, I would not want to prematurely abandon it just so others would be willing to help out. Before I take this radical step, I'll ponder other options.
Obviously, directed panspermia is not an area of fertile research. This is not because the hypothesis has ever been falsified. Most scientists dismiss it for philosophical reasons, citing that it "simply pushes the problem one step back" (I think this argument is really an expression of human psychology). Nevertheless, I think the Crick and Orgel article nicely answers these arguments.
Directed panspermy remains a viable option in explaining life's origin (it even gets a small discussion in a recent Evolution textbook). Front-loading is a hypothesis that extends from it, as I am assuming the planet was seeded for a purpose that goes beyond the original seeding.
You write: In any case, I agree with you on the prospective utility of the front-loading concept, but it seems to me that the problems lie with the retrospective utility of discovering the front-loading effects. In other words, can we in fact reconstruct an initial state, or is it always an assumed entity?..... Correct me if I am mistaken, but it seems like the approach you have taken is to assume an initial state, and then match your observations with regards to possible evolutions of that state.
I agree this is a problem, but it is a problem, to one degree or another, shared by anyone trying to explain history. Consider abiogenesis. You will not find powerful arguments establishing that the earth did indeed spawn life, even in a scientific sense. Clas Blomberg, a researcher of abiogenesis, writes that origin of life research "is aimed to show how it could have happened"(emphasis added/ J Ther Biol 187:541-554). And when it is time to evaluate various speculations about the origin of life, Blomberg points out that the "primary question is not: "is this the way it happened?", but rather, "what arguments support the possibility that it could have occurred this way and what speaks against it?""
Throughout all my front-loading speculations, you will not find me insisting this is the way it happened. You will simply find me speculating about way things could have been designed and considering arguments that support and speak against such possibilities. In this case, since we agree about the prospective utility of front-loading, anyone who wants to give ID some serious thought ought to consider the plausibility of this method of design. That's all I am doing. You note that this approach is very risky, especially when combining assumed initial states with assumed intentions of an agency. I agree. Thus, mark me down as one willing to explore risky territory and simply recognize that I do not expect or demand others to follow along. Remember, this is brainstorming and I already laid out some of the lessons:
quote:
Thus it is important to realize that, compared to OOL research, we expect nothing more from ID. That is, like origins research elsewhere, ID need only generate a coherent, plausible, data-consistent narrative that speculates about how things could have happened. During the investigation, the ID theorist is under no obligation to discard a hypothesis the moment a piece of data "falsifies" (as Andy notes). And while we may never be able to extract definite information about life's design, it is still worth discussing and thinking about.
Of course, this awareness is balanced by the consideration that the skeptic is under no obligation to accept or embrace ID until something very solid is produced (just as no one is rationally obligated to accept or embrace traditional OOL views until something solid is produced).
http://www.iscid.org/boards/ubb-get_topic-f-6-t-000282.html
As for the designers, you note: In other words, the IHE hypothesis makes sense only in light of an engineer that designed for the purpose of minimizing inteventions of his inventions across time. Or put more pointedly, you assume the engineer is not a micromanager.
I think you are correct. In think there is sufficient circumstantial evidence that merits a design inference for life itself. From here, it is possible that no subsequent interventions would occur. However, I am open to the possibility of subsequent interventions (as front-loading may have been setting the stage for such events), but I would rather consider them if a further mass of evidence indicates it. On my web page, I make it clear that I am trying to explore biotic reality through three teleological perspectives: origin events include the intervention of mind at key spots in natural history; design "front-loads" evolution and thus imposes a bias on evolution; and evolutionary mechanisms are themselves the products of rational design. I am currently exploring all three perspectives and not advocating any one in particular.
The purpose of my political analogy was to highlight how humans go about designing the future. Since they cannot control the future as puppet-masters, they rig things that make certain futures more likely than others. This is relevant to your following comments:
OK. Then it seems that this engineer really did not have any specific hidden states in mind with respect to cytosine usage.
It's the degree of specificity that matters. The engineers may very well have rigged things such that multicellular life forms would appear. Or that certain types, rather than other types, of multicellular life forms would appear. But this doesn't mean they had a particular species in mind that would appear at a certain time and place.
After all, if the secondary design need not have a selectable function until much later in time, it makes little sense why the engineer didn't directly implement the design in the first place.
You seed a planet. How do you directly implement the design a billion years later?
Otherwise, if at least one of the selectable function was a necessary intermediate, then he faces a daunting exponential problem of ensuring its arrival. Furthermore, as you mentioned, C->T are not the only mutational events. Are we also considering the possibility of deletions? recombinations? How can this engineer account for all those possibilities while waiting for the states to unpack?
These are problems to be explored, not reasons to abandon the hypothesis. The overall picture of evolution is not this dismal. The diversity of life is only skin deep, many proteins/functions have been maintained, largely unchanged for billions of years, and convergent evolution is not uncommon. For illustrative purposes, I am not proposing that Drosophila melanogaster itself was front-loaded. Instead, perhaps something fly-like was front-loaded.
Moreover, the narrative that you build becomes frustratingly dependent on building a complete picture of this entity's intentions, one which a non-teleologist is free from considering directly. This problem manifests especially when you ask that I suspend judgment on what I perceive as a lack of efficiency but in the same breath suggest that your perceived efficiencies are more believable than current models.
I do not see the problem here. If we are to make an efficiency claim, we need an object of comparison. And compared to a model which advocates an equivalent likelihood of base substitutions that result in amino acid changes where amino acid properties are changed in random directions, the imposed direction entailed in my (incomplete) model appears more efficient. If you have a more efficient way of designing multicellular creatures through single-celled organisms, feel free to share it. If it's simply about a designer that should have intervened directly, you need to explain why such designers could have and would have.
I'd like to know how you are so certain of this bias
I expressed no certainty; I said "there seems" to be a pattern considering the Chou & Fasman rules. Any hypothesis is constrained by current understanding.
Anyway, I'm going to hold off from further discussion of this generic philosophical material (interesting as it might be) and turn my attention to another feature related to cytosine deamination coupled to the code.
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 02. May 2003 21:44
I stumbled upon this in my pile of papers:
Mol Biol Evol 1995 Jul;12(4):527-3
Initial mutational steps toward loss of opsin gene function in cavefish.
Yokoyama S, Meany A, Wilkens H, Yokoyama R.
The river-dwelling fishes of Astyanax fasciatus and their derivatives from Pachon and Micos caves in Mexico were collected. By comparing red (r007) and green (g103) opsin gene sequences from the three populations, we found a high frequency of C-->T changes in r007 and g103 of the eyed Micos fishes presumably due to spontaneous cytosine methylation or enzymatic deamination of the CG dinucleotide, and a deletion of 12 consecutive nucleotides in g101 of the blind Pachon fishes. Thus, high rates of nucleotide substitution and C-->T transitions, usually found for pseudogenes, may occur even before they become nonfunctional.
IP: Logged
|
|
Mesk
Member
Member # 630
|
posted 04. May 2003 22:52
quote:
Mike Gene:
The diversity of life is only skin deep, many proteins/functions have been maintained, largely unchanged for billions of years, and convergent evolution is not uncommon. For illustrative purposes, I am not proposing that Drosophila melanogaster itself was front-loaded. Instead, perhaps something fly-like was front-loaded.
Surely the design would need to be even more general than that for front-loading to be plausible? For instance, cytosine deamination could conceivably have been used to front-load some general bias towards certain structural/chemical properties of proteins, but isn't it rather implausible to suggest that such a mechanism could be used to front-load "tendency to evolve flying insects?"
I understand that cytosine deamination is only one of the mechanisms which you propose may have conveyed front-loaded information over evolutionary time, but I find it hard to imagine that any possible front-loading mechanism could have carried such detailed information as "make something that looks like a fly" - such detailed information would surely have been lost over billions of years of accumulated evolutionary noise. More plausible, IMO, is front-loaded information which provides far more general instructions - perhaps a general trend towards increased genetic adaptability and increased genomic complexity, for instance. Fruitflies, flowers and flying foxes would thus all have been unpredictable (but satisfactory) consequences of such a set of general instructions, rather than predestined outcomes of a detailed front-loaded schematic.
I would be interested to hear your take on this, Mike. How detailed do you think front-loaded information might have been? Could the designer(s) really have specified "fly-like organisms" in their front-loaded blue-print, or are wings and compound eyes merely one of many possible solutions to environmental problems, aided in their development by useful but very general front-loaded mechanisms?
[ 04. May 2003, 22:54: Message edited by: Mesk ]
IP: Logged
|
|
Mike Gene
Member
Member # 149
|
posted 08. May 2003 00:47
Hi Mesk,
You asked: I would be interested to hear your take on this, Mike. How detailed do you think front-loaded information might have been?
I have no idea, as this is the very question I seek to explore. At one level, it might only be a general trend towards increased genetic adaptability and increased genomic complexity, as you suggest. At the other end of the spectrum, it could entail the eventual appearance of some type of sentient being (like us). Currently, I am focused on gathering evidence that argues for the plausibility of some level of front-loading and speculating about mechanisms. I think I have made some progress in these regards. Eventually, I hope to apply such a perspective to the origin of metazoa.
I'll add some more comments when I get time(things are hectic now).
IP: Logged
|
|
|
Printer-friendly view of this topic