|
Author
|
Topic: Selection Acting Directly on Genes
|
warren_bergerson
Member
Member # 262
|
posted 21. September 2002 07:46
I propose for discussion here the following scientific hypothesis generated by design science.
GENETIC SELECTION HYPOTHESIS: The forces or processes of selection responsible for genetic change are primarily or predominantly processes operating directly on genetic material rather than processes acting on the phenotype characteristics produced by genetic material.
Before considering the analysis supporting this hypothesis, it will be useful to note the following items.
1. The Genetic Selection Hypothesis(GSH), like its Darwinian counterpart Natural Selection(NS), is a scientific hypothesis not a fact.
2. I am under the ‘impression’ that the GSH contradicts and is in direct conflict with the NS Hypothesis. However, since there appear to be multiple interpretations of NS, I leave for Darwinian theorists the question of whether a conflict exists.
3. GSH is not based on any new knowledge of genetics nor is it based on what could be considered expert knowledge of the subject of genetics. GSH is based on reevaluating or reinterpreting well known and generally accepted facts from the field of genetics using mathematical analysis techniques taken primarily from the field of actuarial science.
4. GSH is based on what can what will be labeled ‘actuarial mathematics’. While the soundness of actuarial techniques is firmly established, the application of these techniques to the analysis of selection rates is at least in part new. The appropriateness and soundness of the analysis provided here is, obviously, subject to independent review.
To begin, it is taken here as a given that changes in the genetic make up of population can be described in terms of the interaction between processes which (primarily) increase or increment diversity(typically referred to as mutation) and processes which (primarily)reduce variation (typically called selection). Given that changes in the genetic composition of a population can be described in terms of forces of increment and decrement, if follows that the changes in genetic composition can be analyzed using actuarial multiple decrement techniques.
This analysis begins by introducing some notation. Let G represent gene being analyzed and let A1, … ,An denote the possible alleles of G. We now let ‘mtAx’ denote the probability that Ax will be generated by a mutation in generation t. Let ‘stAx’ denote the probability that if Ax occurs in generation t it will be ‘eliminated by selection’. Finally, let ‘ptAx’ denote the number of occurrences of the allele Ax at the beginning of generation t.
The analysis here uses the stationary population ‘assumption’. It is assumed 1)that each mtAx is the same for some extended period of time, (but obviously not for all x), 2) that each selection rate stAx is the same for some extended period of time, (but again obviously not for all x), 3)that the total population is stable, and 4) that the distribution of allele is (relatively) stable (ptAx does not change materially from one generation to the next. The stationary population assumption is a standard actuarial assumption or convention used to simplify calculations. It is suggested here that the stationary population assumption will not distort the results of the analysis. Techniques exist for testing the validity of this assertion.
If a population is stable or (relatively) stationary, and if you have reasonably reliable estimates for the rates of increment (mutation), then you can calculate/estimate the rates of increment(selection) based on the current population(distribution of alleles). This may not be obvious, but it follows, I suggest, from basic actuarial concepts or principles. The validity of those principles can, I believe, be demonstrated mathematically. [The discussion here is based on what can be described as generally accepted actuarial mathematics. While I have no reason to believe that there are any fundamental flaws either in actuarial mathematics or in the way it has been applied, it is important to note that actuarial mathematics and its application here are subject to the appropriate technical review. I interject this cautionary note because the results obtained using actuarial techniques appear to be quite different from the results obtained by other analytical techniques. ]
Since techniques exist which provide ‘reasonable estimates’ for the probabilities of different types of mutations, actuarial techniques can be used to develop ‘reasonable estimates’ for the selection rates applicable to the different alleles of a gene G.
[Let me note here, once again, that the discussion here is based on a limited general knowledge of genetics. It is my understanding 1)that ‘common mutations’ have probabilities of occurrence in the ‘1 per million’ to ‘1 per billion’ range and 2)that for at least some genes, there are relatively high concentrations of a limited number of alleles. The analysis performed here suggests these ‘facts’ are compatible with GSH and not with NS.]
Knowledge of genetics suggest that there are very large numbers of potential or likely mutations. Both phenotype and genotype data suggests that there is a high concentration of ‘common types’. In other words, the distribution of ‘alleles in a population’ is very different than the distribution of alleles which would be generated by known mutation processes. Using standard actuarial techniques, these ‘observed results’ suggest that the selection rates applicable to most alleles (i.e. most sAx’s) are 1)very close to 100% and 2)are equal to 100% for most potential alleles. [I apologize if I am being overly cautious, but I am stating this as an obvious conclusion resulting from well known facts and standard actuarial calculations. I am not showing the calculations because, I claim the result is obvious. If I have made some logical blunder, then my ‘claim’ can be easily refuted by contrary evidence.]
Given that selection rates applicable to most potential alleles are 100% or close to 100%, the question arises ‘What is the process or mechanism producing these very high mutation rates?" Traditional ‘theory’ suggests that a selection rate at or near 100% means the phenotype expression of the allele is either ‘almost always fatal’ or ‘almost always prevents successful reproduction’. But several types of evidence strongly suggest that this is at least very unlikely.
For many species, we have, or can develop reasonably accurate measures of mortality/reproductive success for the period after the time the male and female cells combine. If ‘death’ were to occur when ‘fatal allele’ first became active, then we should observe a very clear and predictable pattern of involving relatively high rates of mortality during the developmental processes. To my knowledge, no such pattern of mortality has been identified, and clearly no such pattern of mortality is exhibited at the latter stage of life.
If natural selection is operating to account for all the high selection rates, it therefore appears that this selection was operating before an egg and sperm combined. But clearly the phenotype traits exhibited by mature organisms are not apparent in the egg and sperm cells. It therefore logically follows that natural selection on mature organism phenotype traits can not be operating on egg and sperm cells.
Having found no evidence that natural selection can account for the high rates of selection ‘observed’ for most potential alleles, we are led to propose that ‘most of the observed high rates of selection involve processes and mechanisms operating directly on genes’. This is the argument/evidence supporting GSH.
The GSH ‘predicts’ the existence of selection processes and mechanisms operating directly on genes. The existence or non-existence of such mechanism therefore provides direct tests of the hypothesis being proposed. While it is not at all difficult to speculate on physical mechanisms that could produce ‘selection operating directly on genes’, such speculation is not necessary to the formulation of GSH. Also, I do not claim to be knowledgeable enough in the field of genetics to identify or describe such mechanisms.
SUMMARY
I have presented here a set of arguments/evidence that justify the formulation of the Genetic Selection Hypothesis. The arguments/evidence are reviewable and refutable at every step. The hypothesis presented produces a number of clearly testable predictions. GSH, it appears, it a perfectly legitimate scientific hypothesis subject to objective review and testing.
The GSH asserts that genetic change is in large part the result of selection processes operating directly on genes. I leave it to others to determine if this is a semantic or substantive difference from existing theory.
No one has complete knowledge of all the subjects relating to the formulation of any scientific theory. Scientific theories are formulated using the knowledge of the person proposing the theory. Other scientists with specialized knowledge in specific areas, then confirm or refute the logic and facts underlying a proposed theory.
As is not uncommon in science, the GSH is the result, not of new knowledge, but of combining knowledge from a number of different areas. In this presentation I have tried to be careful in identifying the limits of my knowledge and expertise in various areas. As should be obvious, the soundness of GSH depends on the soundness of the arguments and the validity of the predictions produced, not on my credentials or lack of credentials.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 21. September 2002 08:12
Despite the claims of obviousness I would like to encourage Warren to support his hypotheses with a more thorough analysis. That's a far more productive approach than inviting others to disprove a vague handwaving hypothesis.I am encouraged to see Warren think of new ideas and welcome any supporting evidence and analysis he may wish to offer.
An interesting discussion on Warren's ideas has taken place on ARN
I would like to invite Warren what effort he has undertaken to show an application of his ideas? My reading of the thread on ARN suggests that such efforts have yet to be undertaken. Yet in order to show any usefulness in Warren's ideas, should one not first attempt to show that they have any relevance to the topic at hand?
Is it not up to the one proposing a hypothesis to provide for supporting evidence?
The ability for it to perhaps falsify Darwinian concepts and theories would surely be ground breaking.
I am looking forward to Warren showing that this is indeed the case. This would truly be revolutionary.
[ 21 September 2002, 09:33: Message edited by: Frances ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 21. September 2002 10:55
Francis,
Please ignore my earlier response.
I will be glad to try to fill in any specific details you may find confusing. One of the reasons for posting a hypotheses in this forum is to identify areas people do not understand and/or do not agree with.
As I attempted to point out in my original post, a significant portion of my analysis is based on actuarial mathematics. In some earlier discussion of this and related topics I did not fully appreciate the extent to which this analysis was based on what are essentially an obscure, and very complex form of mathematics. [After 25 years, you start forgetting how difficult it was to learn certain techniques.]
It is difficult to reduce actuarial mathematics to a level understandable by actuaries let alone non-actuaries. If you can identify specific areas which are unclear, however, I will make an effort to clarify the concepts and techniques involved.
[ 21 September 2002, 12:12: Message edited by: warren_bergerson ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 21. September 2002 12:33
Dear Warren,
There are several questions I have for you to ponder. The first one is: could you explain why you believe that your ideas falsify Darwinian concepts? After all such concepts have been quite succesfully applied in many areas. My question to you is to show through an actual model or through a mathematical derivation support for your thesis. Furthermore it would be interesting to see if your ideas can be applied equally or more succesfully as Darwinian concepts.
I do not think that actuarian mathematics is that obscure or very complex but I have yet to see a clear application of actuarian concepts to your ideas. You throw around some terms which may or may not have any relevance to your claims and ideas. Thus I propose for you to provide us with perhaps a general outline of how these mathematics can be applied to your ideas. Seems to me that a lot of the math is similar to markov chains, optimal control, stochastic processes etc. Certainly in insurance the concepts of population and selection are quite useful but that does not mean that evolutionary processes could be captured using this approach in a meaningful manner. Genetics and actuarian science do overlap mostly in determining life expectancy issues.
But assuming that it could, that by itself does not show that Darwinian processes have been disproven.
Thus my questions to you are simple: Show that actuarian mathematics are relevant to the problem, show that when applied they disprove Darwinian concepts.
So far you have provided few details supporting your model. And you know what they say about "the devil is in the details"...
I also have another observation. You wrote
quote:
I interject this cautionary note because the results obtained using actuarial techniques appear to be quite different from the results obtained by other analytical techniques.
¨
and
quote:
I apologize if I am being overly cautious, but I am stating this as an obvious conclusion resulting from well known facts and standard actuarial calculations.
Does this mean that actuarial techniques have already been applied to biology. I would love to see the details.
Perhaps the actuarian mathematics are a bit more complicated
When you stated on ARN that
quote:
ere is a relatively simple actuarial demonstration (the actuaries I talked to could do the calculations in their heads) that shows that if 1)you assume that either Darwin or neo-Darwin theories are valid, 2)you assign mutation and selection rates consistent with known facts, 3)you assume realistic environmental conditions and 4)you simulate a realistic population with a realistic genome over a period of time, everything dies.
Were you talking about actual calculations. If so, then perhaps you could show these simple calculations based on your 4 assumptions.
I am very interested in hearing more about these calculations.
[ 21 September 2002, 12:44: Message edited by: Frances ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 21. September 2002 17:29
Francis,
Quote: Furthermore it would be interesting to see if your ideas can be applied equally or more succesfully as Darwinian concepts.
I firmly believe that the ‘value’ of a science is in its practical applications. However, the practical applications of design theory versus the practical applications of Darwinian theory is well outside the scope of this thread. A good subject for another day.
Quote: I do not think that actuarian mathematics is that obscure or very complex but I have yet to see a clear application of actuarian concepts to your ideas.
Then you understand are in agreement the techniques that produced the conclusion-
"Using standard actuarial techniques, these ‘observed results’ suggest that the selection rates applicable to most alleles (i.e. most sAx’s) are 1)very close to 100% and 2)are equal to 100% for most potential alleles."?
This conclusion is based on relatively simple measurement of mortality techniques. To claim to understand actuarial mathematics, and then not recognize the validity of simple application of that mathematics is somewhat inconsistent.
I attempted to present the logic leading up to GSH step by step. It would be more productive to the conversation if you addressed the specifics of the argument rather than making generalizations which germane to the specific argument presented.
Quote: But assuming that it could, that by itself does not show that Darwinian processes have been disproven.
If you will reread my original post you will note I specifically avoided the issue of falsifying Darwin.
GSH is, IMO, a potentially interesting and complex topic. I think I would be more productive if at least the initial discussion focused on the subject presented.
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 21. September 2002 19:41
It may be worthwile noting that the gene (as opposed to the organism, the species etc) has been proposed as the one and only focus of natural selection by none other than the darwinian orc Richard Dawkins and his acolites something like 25 years ago (in fact, several of his ideas had already been proposed even earlier).
Still, according to Dawkins' hypothesis, most genes are selected as a result of their phenotypic effects (direct or "extended", in his words), although a class of sequences ("selfish DNA", such as repetitive DNA, transposons and such), may not. Nevertheless, even "selfish DNA" is ultimately subject to natural selection at the organismal phenotype level, once its selfish expansion impinges on the phenotype.
So, my question to warren is, besides his mathematical model (whatever that is), how does selection occur, in real life, on a gene (say, my own personal, entirely novel alpha1-antitrypsin allele) without regard to its phenotypic effect?
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 22. September 2002 02:55
Charlie,
Quote: So, my question to warren is, besides his mathematical model (whatever that is), how does selection occur, in real life, on a gene (say, my own personal, entirely novel alpha1-antitrypsin allele) without regard to its phenotypic effect?
A good, relevant question. In a broad general sense, evolutionary change is ‘driven’ by the by the fact that ‘phenotypic effects’ increase the likelihood of survival. If terms of design concepts, evolutionary and adaptive change is ‘driven’ by the benefits of finding a teleological or functional form Nf from a potential set of N forms. Although the terminology is different, Darwin and design science agree that the end result of evolutionary/adaptive change processes is the functional, adaptive or ‘survival positive phenotypic effect’.
The question being raised is ‘Does nature find Nf in N by randomly looking at Nx and Ny in a very large genetic N, and select Nx over Ny if Nx is more adaptive than Ny’(this is essentially neo-Darwin RM&NS theory). or ‘Is there an alternative, more direct route by which nature searches and finds Nf in N?"
Put in somewhat different terms, "Is evolutionary/adaptive change a process to ‘directly search for Nf in N’ or "Is evolutionary/adaptive change a process to develop efficient search processes which make it possible to quickly find Nf in N?" The GSH suggests that evolutionary/adaptive change is not a direct process. Short term, immediate or direct selection does not act to pick between ‘Nx and Ny’. Short term, immediate, or direct selection processes operate ‘directly on genes’. This in turn creates a powerful ‘search mechanisms’ which indirectly selects between Nx and Ny and eventually locates Nf in N.
The proposal/concept that evolutionary/adaptive change processes operate on ‘evolutionary processes’ rather than on ‘evolutionary effects’ is not new. The analysis underlying GSH is simply a technique for showing that the ‘indirect explanation’ of evolutionary change is more appropriate than the simple direct explanation.
My view is that there exists a big discrepancy between ‘What we know about evolution" and "The theories and models we use to explain evolution’. Darwin and RM&NS, IMO, are theories that science outgrew a long, long time ago. However, because of some rather complex technical issues, no one knew how to formulate rigorous scientific theories capable of fitting the ever expanding body of knowledge. For reasons that say more about human nature than about science, the relationship between ‘evolutionary theory’ and ‘evolutionary knowledge’ became a relationship between ‘a belief system’ and ‘a set of facts’ rather than the relationship between ‘scientific theories’ and ‘supporting evidence’.
On the surface, the suggestion that selection processes and mechanism operate directly on genes, seems to directly contradict a widely accepted Darwinian belief systems. But as a practical matter, we recognize that such selection mechanisms exist (gene ‘repair’, for example, is a process which selects out and eliminates certain types of mutations). We recognize that evolution operates on evolutionary processes. On the basis of existing knowledge, the suggestion that evolution is an indirect rather than a direct process is easy to accept.
It is only, IMO, when the ‘indirect view’ of evolution comes up against Darwinian beliefs, that we get a conflict. The analysis underlying the GSH is a relatively straight forward technical analysis using validated analytical techniques (although apparently not techniques validated in EB). If someone would actually follow through and review the technical analysis, I am confident they would confirm my findings. They would also find that my findings and conclusions are compatible with what we understand about how evolution works. The only ‘conflict’ created by my analysis, IMO, is with the Darwinian belief system.
In conclusion, I think you asked a very good question. The answer is "selection, in real life, on acts directly on genes ((I don’t know about the specific example)say, my own personal, entirely novel alpha1-antitrypsin allele) and which is an indirect result of ‘regard for its phenotypic effect’. The hypothesis I propose simply suggests that ‘evolution operating on evolutionary processes’ is a better description than ‘evolution operating directly to produce ‘phenotype effects’.
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 22. September 2002 08:26
I don't think you answered my question, warren (or at least I can't find the answer in your post). I asked: how does your method of selection at the level of gene differ from Dawkins', and if it does, how is it effected.
I can put two populations with slightly different genotypes in a certain environment, and observe natural selection in action, in darwinian terms. I can go back and measure fitness of individual alleles, and with some more effort even figure out how they affect fitness. How do you observe selection at the level of genes in a similar example? Would the end point differ from that expected from darwinian models, and how? And, most importantly, what does this novel method of selection consist of, beside mathematics? In other words, what natural (or supernatural) "force" effects the selection, on what basis, and how, if not by changes in gene frequencies based on differential reproductive success ?
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 22. September 2002 09:43
Warren,
I am still looking forward to these simple calculations which seem to disprove the Darwinian concepts.
You make claims that remain unsupported by any supporting calculations or mathematics. When I tell you that actuarian mathematics is hardly that obscure or complex you jump to the claim that thus I should understand your unsupported claims about genetics and actuarian mathematics. Could you please describe these 'standard actuarian techniques' that led you to jump to this conclusion?
I was under the impression that you had done the actual mathematics? Are you now suggesting that your claims need yet to be supported by actual mathematical theory and calculations ?
Since you want to focus on GSH I would invite you to provide us these simple calaculations which lead to your conclusions?
That should not be too much to ask for since you seem to suggest that such calculations have already been done. Providing us with these calculations could surely provide some necessary and so far lacking support for your claims.
You claim that you have reached certain findings that could be recreated by others using the same techniques. Surely this seems to suggest that you have such calculations performed and thus I invite you, in order to further this dicussion, to provide us with them. Otherwise we may have to conclude that it is to early to seriously consider your ídeas and have to wait for a time in which a more solid foundation can be provided.
Looking forward to your contributions of your calculations.
[ 22 September 2002, 09:46: Message edited by: Frances ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 22. September 2002 14:13
Charlie,
I interpreted your original question as ‘How do explain selection acting directly on genes, when the ‘result’ of the selection is to produce phenotype effects which are adaptive or which increase the likelihood of survival?" That is the question I addressed. Sorry if it was not the question you asked. The more specific questions you presented in your follow up are even better than the general question in your earlier post (IMO). I have attempted to answer these questions in terms of specific analysis that has or could be performed. If you have questions relating to specific analytical processes, I will be glad to discuss them in more detail.
Quote: How do you observe selection at the level of genes in a similar example?
If you have 1)a reasonable estimate of mutation rates for different types of mutations and 2)you know the current distribution of alleles in a population then, using what I believe are standard actuarial techniques, you can develop reasonable estimates of the applicable selection forces or rates.
Given estimates of selection rates, you can work backwards, by a process of elimination, to determine if the selection process producing the measured rate of selection is due to or explained by natural selection- selection based on differential survival and/or reproduction operating on ‘phenotype effects’.
GSH does not deny the existence of ‘natural selection’. GSH simply asserts that the level or magnitude of the force of selection observed/measured in biological systems can not be explained by natural selection processes. Demonstrations that natural selection can not explain most of the observed ‘forces of selection’ could be produced, I suggest, using standard actuarial and analytical techniques.
Quote: Would the end point differ from that expected from darwinian models, and how?
Given a search space N, a set of adaptive solutions Nf, and some starting point Nx a member of N but not a member of Nf, then the average or expected speed of finding a member of Nf will depend on the rate of selection. The higher the rate or force of selection the faster the speed of finding Nf. I believe this has been confirmed by analysis with GA models.
The GSF says forces of selection operating on genes involve a combination of natural selection and ‘selection processes other than natural selection’. The GSH predicts that in some or most instances ‘selection processes other than natural selection’ involve significantly greater forces of selection than can be generated/explained by natural selection. This in turn suggests that the total forces of selection are greater than the forces of selection due to natural selection alone.
The GSF thus predicts that evolutionary change occurring in biological systems will be faster than would occur if evolution involved only Darwinian natural selection. [Although not addressed here specifically, design theory asserts that survival depends of ‘adaptive or evolutionary’ speed as well as the occurrence of evolutionary change]
Quote: And, most importantly, what does this novel method of selection consist of, beside mathematics? In other words, what natural (or supernatural) "force" effects the selection, on what basis, and how, if not by changes in gene frequencies based on differential reproductive success ?
As I stated in my original post, GSH is based on techniques which measure(appear to measure) the existence of forces of selection other than Darwinian natural selection. While the GSH ‘predicts’ the existence of natural physical-chemical processes responsible for non-Darwinian selection, the formulation of the theory is not based on the identification of these natural mechanisms. [IMO, this is not unlike ‘predicting’ the existence of an as yet unknown planet based on observed unexplained distortions in the observed orbits of known planets. ]
Although, as I stated earlier, I have no expertise in identifying the physical processes responsible for non-Darwinian selection, it is easy to speculate on the possible nature of such mechanism. As a simple example, consider a mutation process which produces 10 different mutations. Now consider a ‘gene repair mechanism’ which repairs or selects out 9 of the possible mutations. Such a repair mechanism is an example of a non-Darwinian selection mechanism.
Quote: I asked: how does your method of selection at the level of gene differ from Dawkins', and if it does, how is it effected.
It seems, at least on the surface, that GSH is significantly different than Dawkins hypothesis. However, as I stated in my original post, for the discussion here, I prefer not to make any assertions about whether GSH is consistent with or incompatible with any existing models or theories.
Again, let me thank you for raising specific and relevant issues.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 23. September 2002 04:55
Warren:
quote:
GSH does not deny the existence of ‘natural selection’. GSH simply asserts that the level or magnitude of the force of selection observed/measured in biological systems can not be explained by natural selection processes. Demonstrations that natural selection can not explain most of the observed ‘forces of selection’ could be produced, I suggest, using standard actuarial and analytical techniques.
I hope that Warren understands that an unsupported assertion such as 'the level or magnitude of the force of selection observed/measured in biological systems can not be explained by natural selection processes' are meaningless unless one can provide for the actual, and at this moment still missing, calculations.
I am still patiently awaiting Warren's efforts to do such calculations and present them to us. Lacking such calculations we have nothing more than a meaningless speculation about the forces of selection and mutation.
It does surprise me a little to see Warren jump to such conclusions without a foundation in modeling, actual calculations etc. Given Warren's claims on one hand they seem to suggest that such calculations have been performed but given the lack of any evidence of such I remain a bit confused.
Warren, when you made these claims about selection did you have any such calculations available or were they mere speculations for which you are now searching for supporting evidence?
And while we are discussing claims, could Warren explain why repair mechanisms are non-Darwinian? Is this based on a thorough analysis of is this a mere speculation on the part of Warren?
[ 23. September 2002, 04:57: Message edited by: Frances ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 23. September 2002 05:28
GSH and neo-Darwinian genetic theory appear to suggest very different and contradictory conclusions. Darwinian theory appears to suggest that the primary source of selection in evolutionary and genetic change is ‘Darwinian natural selection’. GSH, by contrast suggests that the primary source of selection in genetic change is ‘non-Darwinian selection acting directly on genes’. It may help clarify the issue to look at 1)why design science produces conclusions so different from and EB, 2) the issue Francis keeps raising, the evidence supporting the different conclusions, and 3)the significance of the different conclusions.
WHY SUCH DIFFERENT RESULTS
The primary reason for the very different results is the ‘perspective’ used to view selection. Darwinian ‘theory’ views selection as a single ‘force’ operating on the entire organism. Design science, by contrast, defines ‘selection associated with genetic change’ as a set of forces, with a separate ‘force of selection’ operating on each allele.
In considering the two perspectives, it will first be noted that there is no inherently right or wrong way of viewing selection. The ‘better’ way of viewing an issue is the perspective that produces the more useful result. The ‘single force’ perspective is obviously simpler, and if it produces appropriate results, it is preferable. The ‘force per allele’ approach is more ‘accurate’, and if the approach can be made practical, and if it produces better results than the simple perspective, it is preferable.
Defining selection as a single force means that selection can be defined in terms of ‘results’ or ‘phenotypic effects’. Selection as a single force can be defined in terms of the adaptive results produced by selection. [Although the value/validity of this approach is widely accepted it does present a number of problems that have not been resolved. ]
Defining genetic selection as ‘forces operating on alleles’ has the somewhat unexpected benefit of making it possible to quantify the individual ‘forces of selection’ operating on different alleles. The measurement of these forces of selection are based on a standard actuarial technique. Knowing the magnitude of the forces of selection, it is possible, again using standard analytical techniques to back track to the mechanism producing the observed rates of selection.
Design science produces a very different description of the operation of selection in genetic change because it views selection in terms of a set of small, measurable forces. Neo-Darwinian theory, by contrast, views selection in terms of a single aggregate approximation which is not quantifiable.
EVIDENCE SUPPORTING THE TWO APPROACHES
EVIDENCE FOR GSH
The body of evidence supporting GSH appears very clear and unambiguous, and is readily available for verification. The step by step evidence is:
1. Known rates and processes of mutation suggest mutation processes can produce large numbers of different alleles.
2. Known distributions of alleles for specific genes suggest high concentrations of a limited number of different alleles.
3. Using standard actuarial techniques, 1 and 2 suggest that the rates of selection of most alleles are at or close to 100%
4. If selection operates on ‘phenotype effects’ then this selection must occur after the phenotype effect has been exhibited.
5. If all selection takes the form of natural selection, then known rates of mutation combined with measured rates of selection would suggest a clear pattern of mortality/infertility after phenotype effects emerge
6. No such pattern of mortality has occurred.
7. Therefore much or most of the selection associated with genetic change must be due to mechanisms other than natural selection.
The evidence supporting GSH appears clear, verifiable and available for review and testing. Note the evidence is not designed to prove GSH, but simply to provide a reasonable basis for proposing GSH. If Francis or anyone else believes the evidence inadequate, then it seems appropriate that they identify what part of the evidence they consider to be inadequate.
EVIDENCE FOR NEO-DARWINIAN APPROACH
There is, it appears, clear evidence that 1)Darwinian natural selection processes exist, and 2)natural selection can produce genetic changes.
There does not appear, at least to my knowledge, to be any evidence supporting the suggestion that natural selection is the only type of selection impacting genetic change, nor is there evidence natural selection can explain all the selection associated with genetic change.
THE SIGNIFICANCE OF NON-DARWINIAN SELECTION
Evolutionary or adaptive genetic change is, in design science terminology, a change from some Nx not a member of Nf to Ny a member of Nf where Nx and Ny are elements of N and Nf is a functional subset of N.
It is generally recognized that the change from Nx to Ny involves some variation of the standard teleological process involving variation and selection. Darwinian and neo-Darwinian theory suggest the evolutionary adaptive changes are the result of ‘variation or changes from Nx to Nxv’ and ‘selection based on a process or mechanism called natural selection’.
Showing that selection operating on alleles takes forms other than natural selection, suggests that process of evolutionary/adaptive change is, or may be, more complex, more powerful and more indirect, than the simplistic/direct approach suggested by Darwinian and neo-Darwinian models.
Abandoning the simplistic Darwinian and neo-Darwinian models does not mean abandoning evolution. Abandoning the simplistic models, and GSH suggests such models can and should be abandoned, opens the door to far more powerful and effective models of evolutionary and adaptive change.
SUMMARY
As I stated in the original posting, GSH is based on a new, more precise and more productive approach to interpreting available facts. The evidence supporting GSH appears clear and readily verifiable/refutable. It is not clear that there is any evidence supporting the claim ‘natural selection is the only form of selection operating to produce genetic change’. It is not even certain if Darwinian theory precludes non-Darwinian selection.
The significance/relevance of GSH, is not that it ‘disproves Darwin’ or ‘disproves evolution’(it clearly doesn’t do the latter and it is not even certain it does the former). The significance of GSH is that it ‘opens the door’ for more complex and more powerful indirect explanations of genetic change. Behe, Dembski and others have demonstrated that simple evolutionary processes can not explain the complexity observed in biological systems. GSH provides evidence that evolutionary change need not be limited to simple Darwinian models.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 23. September 2002 05:45
Warren,
your claims that
quote:
GSH and neo-Darwinian genetic theory appear to suggest very different and contradictory conclusions. Darwinian theory appears to suggest that the primary source of selection in evolutionary and genetic change is ‘Darwinian natural selection’. GSH, by contrast suggests that the primary source of selection in genetic change is ‘non-Darwinian selection acting directly on genes’.
seem to be too premature to be considered valid. You have yet to show that GSH reaches these conclusions based on actual calculations or modeling. You seem to be jumping to a conclusion without the intermediate step of supporting evidence.
For instance your claim about Darwinism views selection as a single force requires some supporting evidence since I am simply suggesting that you are wrong here.
Your 'evidence for GSH'is based on mere speculation, some unwarranted assumptions thus your conclusion remains unfounded.
For instance your step 3. remains unsupported by any evidence.
You fail to provide any evidence of 5.
Warren's statement about natural selection being the only factor in evolution seems to be self fulfilling and obvious. Science has reached that conclusion decades ago.
So before Warren jumps to an unwarranted conclusion that GSH is based on a more precise and more productive approach, he should first make an effort to show that GSH can be applied and that GSH disproves Darwinian mechanisms. Until then we should realize that Warren's ideas are mere speculations without much supporting evidence.
If GSH provides evidence then I am sure that Warren can share this evidence with us. So far nothing of the kind has been provided.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 23. September 2002 08:15
Warren writes,
quote:
The body of evidence supporting GSH appears very clear and unambiguous, and is readily available for verification. The step by step evidence is:
1. Known rates and processes of mutation suggest mutation processes can produce large numbers of different alleles.
2. Known distributions of alleles for specific genes suggest high concentrations of a limited number of different alleles.
3. Using standard actuarial techniques, 1 and 2 suggest that the rates of selection of most alleles are at or close to 100%
Warren, the word “suggests” is fairly mild in comparison with the phrase “very clear and unambiguous, and is readily available for verification.”
Let us assume that someone wanted to look at the data (known rates, known distributions, etc.) so they could verify this evidence. Where would they look. Could you provide a reference to, or summary of, actual numbers based on actual biological research? If this information is “readily available for verification”, I think it would strengthen your argument if you could refer to it. Stating that unspecified “known” rates, processes, distributions “suggest” a conclusion is a statement that others are unlikely to assent to on face value.
My same point applies to statement 3: can you describe these standard actuarial techniques, with at the very least numbers chosen for the sake of illustration. You say that actuarial methods are complex, and hard for people to understand who are not trained in them. But I am sure that there are people here at ISCID that can understand. Many of us (and I count myself as one who has programmed simple evolutionary simulations) understand the idea of iterative processes where both incremental and decremental changes are occuring.
So, again, I believe that even though you say that the evidence is “very clear and unambiguous, and is readily available for verification,” your assertions are too general, and the evidence you offer to support them is too lacking, for us to evaluate them.
It is not more logical analysis from you that is needed, or more definitions and acronyms. It is more (dare I say the word) specificity.
IP: Logged
|
|
peroxisome
unregistered
|
posted 23. September 2002 13:15
quote:
5. If all selection takes the form of natural selection, then known rates of mutation combined with measured rates of selection would suggest a clear pattern of mortality/infertility after phenotype effects emerge
err, no.
Some genotypes clearly lead to pathology or infertility; but there are a whole range of genotypes and phenotypes which merely lead to reproductive success. This busts 5 above, without even starting to consider the effects of mutation operating on a population. I really don't understand where (5) above comes from, because first year undergraduate textbooks on evolutionary theory cover this area clearly enough, and there are several popular works (e.g. Dawkins work) which cover this in detail. This seems to be a straw man argument.
yours
per
IP: Logged
|
|
|
Topic Closed
Printer-friendly view of this topic
