|
Author
|
Topic: Royal Truman: Avida, a biologically unrealistic model for neo-Darwinian Theory
|
Moderator
Administrator
Member # 1
|
posted 23. January 2004 15:50
Note: This file is being preserved for historical reasons only. An updated version of this paper has been posted in its own thread.
Avida, a biologically unrealistic model for neo-Darwinian Theory
by Royal Truman
Abstract: Avida(1) is a computer program designed to simulate how neo-Darwinian evolution could develop novel cellular networks. However, examination of the model assumptions and parameter values raises severe questions as to its relevance to biological organisms.
To read the paper, click here.
[ 01. July 2004, 14:24: Message edited by: Moderator ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 23. January 2004 17:55
Having just counted no fewer than five errors of fact and description (not to mention errors of interpretation) on just the first page of the ms., I don't see that it is worth spending a whole lot of time and effort on.
RBH
[ 23. January 2004, 18:04: Message edited by: RBH ]
IP: Logged
|
|
Moderator
Administrator
Member # 1
|
posted 23. January 2004 19:45
RBH. Truman actually submitted this paper with the explicit hope of getting suggestions and criticism on details similar to what you suggest. I think he would be happy if you would point out a few of these errors of fact and description.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 23. January 2004 23:10
Moderator,
I see. But given the plethora of problems with the draft, starting in the first sentence of the abstract (avida is not "... designed to simulate how neo-Darwinian evolution could develop novel cellular networks"), it's not even at an acceptable undergraduate level. Had an undergraduate student turned this in to me with a request for comments, I'd have made exactly the same remark that I made in my first posting, perhaps adding, "Go read the relevant papers."
Those papers are readily available, and I see no reason to take a Ph.D. by the hand to lead him through the relevant literature. Were I to identify and correct all the errors of fact, interpretation, and inference in the draft, I would require credit as co-author, probably first author (not to mention that correcting those errors would leave the draft bereft of its central argument). As I said, it's not worth my time and effort.
RBH
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 24. January 2004 01:45
The central message doesn't last very long at all. Avida isn't meant to be biologically realistic! It's supposed to (from Lenski et al.'s abstract!) "show how complex function can originate by random mutation and natural selection".
Sure, the authors make a few justifications of why their model isn't completely irrelevant because of the differences from biological systems. But Lenski et al. explain quite clearly in their introduction that they are using this because, "By using this tractable system, we aim to shed light on principles relevant to any evolving system."
Thus, much of the paper is really taking aim at a straw man.
And then there are further problems. A sampling: why does it matter if the Avida genome is miniscule? It's a simplified model--it's supposed to be small! Many organisms have lots and lots of largely or entirely nonfunctional DNA, so it's a better simple model of those than of Mycoplasma.
Likewise, mutational rates are to make computation time tractable and don't really affect the qualitative outcome.
(Yes, if you have weak mutation pressure and very strong selection for very slightly smaller genomes, then your genome is unlikely to expand in size. However, that isn't the case for the bulk of extant organisms on the planet.)
And critically important functions for life aren't that hard to destroy in the original (30% chance per mutation!). Some of the evolved critters are more redundant, so it's harder to wipe them out in one step. This is actually a suggestive result: some aspects of redundancy in biological systems may be the result of hijacking of components to perform a second, nonessential function that, nonetheless, causes redundancy in a system that is essential.
And so on.
The analysis of pressure towards smaller genomes is somewhat interesting, but it would be better presented as a statement of conditions under which genome expansion is difficult.
There are many organisms in which expansion is not a problem (corn, for instance--what a nightmare for genomics work!). For those where it is, Avida would need modification. Evolution via drift of small genomes plus mixing of genetic material, perhaps?
Avida's not a final answer, but as far as limited and partial answers go, it's not half bad.
[ 24. January 2004, 01:52: Message edited by: Rex Kerr ]
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 24. January 2004 23:47
From Avida website we read
quote:
Avida is an auto-adaptive genetic system designed primarily for use as a platform in Digital or Artificial Life research. In lay terms, Avida is a digital world in which simple computer programs mutate and evolve
Avida allows us to study questions and perform experiments in evolutionalry dynamics and theoretical biology that are intractable in real biological system.
quote:
"... designed to simulate how neo-Darwinian evolution could develop novel cellular networks"
More on Avida
quote:
We have developed a computational system, the Avida software, which can be used to study certain basic properties of simple living systems, namely those that do not depend on the particular embodiment of information storage and machinery.
...
We use 'avidians' as a model organism for experimental evolution. While the genome of standard avidians is much shorter than that of the simplest bacteria, their evolutionary dynamics is quite similar. In collaboration with Richard Lenski's Microbial Evolution Laboratory at Michigan State University, we are testing evolutionary hypotheses and study their universality across different forms of life. At the same time, we are developing new hypotheses based on data obtained both with biochemical, and digital organisms
Or Lenksi
quote:
‘Our goal is not to mimic natural systems in detail, but rather to expand Avida to give digital organisms access to more of the basic processes of life', says Lenski. ‘Our goal is not so much to endow the ancestral organisms with additional capabilities, but rather we want to see how digital organisms will evolve if they are placed in an altered world where such things as sex and communication are physically possible. I see many years of interesting research along these lines'.
Digital Evolution Bill O'Neill, Digital Evolution
Bill O'Neill PLoS Biol. 2003 October; 1 (1): e18
Or Adami
quote:
"They are literally alive, only it's a foreign type of life," says the Caltech scientist, physicist Christoph Adami, who teaches a class called "Artificial Life" at the Pasadena campus. He is one of five authors of the Nature article.
Digital critters mimic behavior of real life
Researchers use them to test Darwin's theories
Keay Davidson, Chronicle Science Writer Thursday, July 19, 2001
[ 24. January 2004, 23:52: Message edited by: Pim van Meurs ]
IP: Logged
|
|
Salvador T. Cordova
Member
Member # 959
|
posted 28. January 2004 11:55
Dr Truman,
If you're reading, I would have you know I found something amusing in the exchange of public emails between Tim Wallace and Tom Schneider:
quote:
Tim Wallace asked:
Do you know Dr. Royal Truman? His article “The Problem of Information” seems to indicate that he does indeed agree with the statement. He is a biochemist.
quote:
Dr. Schneider Responded:
Yes, I’m familiar with "The Problem of Information" It has lots of flaws, but you might like to know that he is basically right, his question has not been properly answered.
I will do my best to personally explore the Avida question over the next few months in support of your paper. I feel the analysis will be valuable in as much as other 'Avida' simulations will inevitably pop up and will be used to justify various evolutionary theories.
I myself believe in evolutionary processes as a part but not the whole of biology. I believe, modelling accurately accurately is important. Avida has been used to justify Darwinian models which claim information increase and creation of irreducible complexity and thus have bearing on ID claims.
My apologies in advance for the delays I may have in offering contributions to your project. I believe something of this magnitude will not be a casual exploration, but will demand exacting attention to technical detail.
On balance, my intuition is that you are right, Dr. Truman, especially given the generous 'backhanded compliment' Schneider gave you concerning the "problem of information".
I hope you will receive our constructive criticism graciously, even if we nit pick. My aim is to help evolve your paper for the better. ISCID is afterall an ID friendly organization, and members like myself wish to do everything we can to promote exploration of the ID hypothesis.
Cordially,
Salvador
[ 28. January 2004, 19:35: Message edited by: Salvador T. Cordova ]
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 28. January 2004 12:07
Lets not give the impression that Schneider agreed with Truman's analysis shall we?
quote:
2001 November 10. The Problem of Information for the Theory of Evolution Has Tom Schneider Really Solved It? was posted by Royal Truman. Thanks to a number of people for pointing out this web page. I will not be able to respond to it in detail in November. For now, I leave it to readers to identify the numerous errors and confusions. An Analysis of Truman's Criticism of the Ev Model will be upgraded when I have time.
The quote referenced by Salvador was about Truman's comments on Dawkins.
But rather than use quote mining here, lets explore some of the identified errors by the participants of this forum.
[ 28. January 2004, 12:10: Message edited by: Pim van Meurs ]
IP: Logged
|
|
Royal
Member
Member # 1060
|
posted 29. January 2004 12:51
Comments to Rex Kerr:
The first line of the abstract (ref. 1) reads: “A long-standing challenge to evolutionary theory has been whether it can explain the origin of complex organismal features”. This is followed by the claim: “These findings show how complex functions can originate by random mutation and natural selection.” To be true, it must be possible to map the Avida model characteristics and extrapolate the published parameter settings to living organisms. I maintain this was not done. I further argued that introducing biological realism would prevent the equivalent of the author’s “logic functions” from evolving.
Intuition: Suppose someone developed a computer model to help you decide whether to build a petrochemical plant. Would this project be profitable? The model must take into account at least the major costs & the cash generated during the plant’s lifetime and the cost of capital. The model needs not be perfect. One could play with various parameter settings to see whether the conclusion of profitability is sensitive to small changes in assumptions.
Suppose you examined the model, found your contractor failed to capture essential internal relationships. Worse, product prices were assumed which are orders of magnitude above market realities. The overly optimistic cash flows prejudice towards assuming the project would surely be profitable. You look deeper, and discover many cost factors (maintenance, insurance, etc.) were completed ignored. Would you agree if someone claimed the model demonstrates this would likely be a profitable business venture? You are left uncertain: do the costs or cash flows dominate?
The details matter very much: net gains or net losses could result. Now, what is the NET result of the kinds of mutations described in the paper in ref. (1)? Populations with more complex functions, quantitatively and qualitatively, as proposed? Or would such populations slowly degrade: lose complex functions and work less efficiently? The model details and parameter assumptions matter very much. One need not work with values, such as mutation rates, which are accurate. But the relationships between key factors, their RELATIVE PROPORTIONS, must be either relevant, or computer runs under several settings must be repeated to permit extrapolation to the real biological world.
For example, it is the nature of the computer instructions used by Avida which ensure that a statistically significant number (in terms of the 3,600 member population) of logic functions will be present among short, random sequences. However, the proportion of random polypeptide sequences able to fold reliably and properly, endowed with the ability to provide novel cellular functions, is orders and orders and orders of magnitude lower. Recall that a metabolic reactions involves several proteins which work together as a well-meshed machine: natural selection can’t fine tune one member, then evolve the next member millions of years later.
The evolutionary framework requires that at some point in the distant past only very small genomes would have existed. Testing very small Avida “organisms” is consistent with this assumption and computationally convenient. But we need to evaluate the true reproductive and putative evolutionary behavior of small organisms under assumptions of a Darwinian, Designer-free, nature. For example, to provide evolutionary playing material, the genomes of the Avida creatures consist of a very large PROPORTION of non-essential genetic material, from which “logic functions” should evolve.
Here the details have to be examined realistically. Avida’s organisms replicate thanks to external software and hardware which provide key services; the energy is supplied from external sources (e.g., electricity); their immediate environment is maintained and repaired by humans. Notice that these requirements cannot be damaged by mutations in the Avida model. However, free-living organisms need to have these critical functions encoded on their genes and these CAN be damaged by mutations. And it is an experimental fact and common sense that streamlined genomes for small, rapidly reproducing organisms, will outpopulate the members carrying superfluous genetic material.
The net effect, when one uses realistic proportions of parameter settings in view of the above comments, in an evolutionary scenario, leads surely to a very different conclusion than reported for the Avida runs (ref. 1). Lets reason it out stepwise:
- The number of random mutations which need to be generated before a truly novel (biological) cellular function is stumbled upon is monstrous
- The amount of indispensible coding material to ensure all key functions are maintained is large (say, 250 genes or so).
- This implies that the proportion of random mutations which are deleterious or worthless is much larger than assumed in the paper
- The reasoning so far demands an incomparably greater number of generations to provide the statistical opportunities to stumble on useful mutations
- However, removal of unneeded genetic material would have a positive selective value: the more generations needed to discover useful mutations, the more opportunities for streamlining (deletions) to occur
- We are left with no, or too short, genetic material to create novel functions long before useful mutations are generated
- Life goes on. Additional mutations which are not deadly would not all be weeded out. The proportion of slightly deleterious mutations is vastly greater than slightly better. As more and more members survive with less-tuned cellular functions, natural selection weakens on average. This permits more slighly deleterious mutations to fix.
These facts are camouflaged by the assumptions, which are design to ensure the outcome reported: the simulation starts off with a miniscule proportion of the genome able to be damaged by mutations; GENOME ENLARGEMENT IS REWARDED even in the absence of new “logic functions”; the Avida genomic sequence space is loaded with evolutionary stepping stones leading to ever more complex logic functions; new logic functions are rewarded with dramatic and selective advantages.
Only the careful selection of model characteristics and the interplay of unrealistic parameter settings permitted the Avida system to develop new logic functions. The relevance to the origin of biological life on earth is absent.
IP: Logged
|
|
Royal
Member
Member # 1060
|
posted 29. January 2004 13:12
About Dr. Tom Schneider: I have read most, perhaps all, his papers, with much interest. In particular the demonstration that the quantitative information content (in Claude Shannon’s sense) permits binding sites to be placed in the right number on the genome and to be identified unambigiously.
If he chose to accept the Intelligent Design view of where life came from, he would have some very interesting things to say about his own work! He believes novel binding sites evolved from random sequences. I am convinced most were designed. In fact, I propose they were originally created with more information content than needed, to provide fault tolerance towards mutations. I suspect the Design also included the notion of how long proteins should bind to specific sites, as part of the regulatory strategy.
Dr. Schneider apparently approved of at least some of the things I suggested on an on-line essay (http://www.trueorigin.org/dawkinfo.asp) in which Dr. Richard Dawkins was on the receiving end, but was not quite so pleased when his own ideas received some of my friendly criticism: http://www.trueorigin.org/schneider.asp
I admit I got a little carried away in the last essay. But the central challenge remains unanswered: without the necessary sensitivity analysis he is not justified in making any claims as to biological relevance. He has the software and it would require very little effort for him to plunk in the parameter settings I suggested and to then report how his critters “evolve”. The proposed simulation settings are biologically-speaking very generous on my part, and are actually unrealistic, but would permit a crude extrapolation.
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 29. January 2004 22:39
Royal's analysis of the abstract jumps from the first line to the last. The intervening lines explain that they were addressing not solving the first issue, describe what they did, and then conclude with what problem they did solve. This format is extraordinarily common in abstracts, and shouldn't be a source of confusion to anyone who reads them regularly (e.g. evolutionary biologists). Lenski et al. therefore are not claiming that they explain the origin of complex organismal features, just as abstracts that begin, "Alzheimer's disease is a debilitating neurodegenerative disease that affects (large percentage) of people over (oldish age)", the paper is not claiming to have cured Alzheimer's.
I agree that details matter, and that relative rates are what's important. Therefore, errors such as calculating 750 point mutations and 600 indels in Mycoplasma per generation--not rescaling for genome length--should be fixed in the critique. (Once fixed, they're not problematic.)
At some point, perhaps only very small genomes existed--but is Avida a model of this? That's not what the article was about. Avida isn't intended to simulate abiogenesis, and the authors make no claims that it is, so criticizing it for not doing this is silly. (Not to mention that the criticism seems to assume without argument that "very small genomes" must have been very fragile to mutation; I would think they'd be robust to mutation but inefficient and fragile to environmental variation.)
The Avida environment is one with constant energy flux, and in such environments it is not the case that the rapidly reproducing organisms are the most successful. They colonize rapidly, but can be supplanted by a larger number of tougher and/or more efficient slower-growing organisms. (Take trees vs. grass, for example.) So while I agree that the unchecked exponential growth model is problematic for evolution of complex functions in the manner that Avida does it, I don't agree that this is any more "biologically realistic" than Avida's conditions (i.e. genome length is irrelevant). They are two extremes, and we observe both extremes in nature as well as everywhere in between.
Note: I don't think it's correct that genome enlargement is rewarded; it simply isn't penalized. What is rewarded is having a more efficient copy loop and executing more logic functions. (It takes more instruction-executions to run and copy a longer orgamism.) If you have evidence to the contrary, please supply the evidence in place of the claim.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 30. January 2004 00:51
There's a misconception woven through Truman's remarks that is explicit in the abstract of his piece. The abstract reads quote:
Avida is a computer program designed to simulate how neo-Darwinian evolution could develop novel cellular networks. However, examination of the model assumptions and parameter values raises severe questions as to its relevance to biological organisms.
The first sentence contains two* errors. First, Avida was not designed to "simulate" anything. Second, Avida has nothing to with "cellular networks," novel or otherwise.
Let me just discuss the 'simulation' point. In a description of the design of Avida, The Biology of Digital Organisms, its designers tell us quote:
The main focus of digital life research is a sort of comparative biology, which attempts to extricate those aspects of simple living systems that are germane to the type of chemistry used, from those that are not [46]. Additionally, digital life can help to refine mathematical theories and aid in developing and quickly testing new hypotheses about ecological and evolutionary processes.
That is consistent with other remarks by the designers. What they mean is that Avida is not a "simulation" of the carbon-based biology of plants or animals or bacteria; Avida is an instantiation of a different system in which (some) evolutionary processes occur. Their goal is to provide a platform in which to do comparative research, contrasting and comparing the evolution of digital critters and biological critters, in order to study general principles of evolution; to test hypotheses about evolution; and (because of the massive data-gathering capabilities) to allow detailed analyses of the time course of evolving systems. We are currently stuck with just one instance of an evolving system, biological life on earth, and Avida is a good first stride on the path to building another system in which evolution occurs, so one can do actual comparative work to test generalizations about evolutionary processes independent of the physical system in which they occur.
I work with computer-based evolutionary processes every day, and I don't deceive myself that they are "simulating" biology; they aren't. Nevertheless I see phenomena that are at least superficially parallel to those one sees in biological systems - differentiation of subpopulations comparable to speciation, variable rates of adaptation (and sometimes extinction) in environments with varying dynamics, the emergence of complicated 'genetic' interactions comparable to epistasis, and so on. Since I work in an applied context with a very tightly focused set of goals for our algorithms, I don't study them systematically in order to draw generalizations about evolution, but I sure see them. The Avida group and others are actually studying them, and the reference above describes some of that comparative research.
So Avida isn't a "simulation" of anything; it is an evolutionary system in itself. It shares some properties with biological systems and doesn't share others. That is precisely what allows one to do comparative research on general properties of evolving systems.
Truman wrote quote:
I further argued that introducing biological realism would prevent the equivalent of the author's "logic functions" from evolving.
Avida, including the source code, is freely available on the Web. Rather than argue it, why not show it? The necessary resources are available.
The Nature paper of Lenski, et al, demonstrated a set of conditions in which evolutionary mechanisms generate products that meet Behe's definition of irreducible complexity, so one can no longer argue that those mechanisms are in principle not capable of doing so. In order to argue that biological "IC" structures cannot evolve by those mechanisms, one must now demonstrate (not merely assert) it; the 'in principle' argument is dead. The platforms in which that test can be performed are becoming available. My bet is that particular ID-of-the-gaps assertion will vaporize as that research is done.
I've looked briefly and informally at a few of the questions Truman raises (mutation rate being one) in a version of Avida I have installed on a fast PC. I don't have the leisure to do systematic research now, though I'm coaching a couple of undergraduate biology majors at a nearby college who may do the worst of the work, and it's not real difficult. While those undergraduates have access to a Beowulf cluster on which we've installed Avida, one can do it, albeit more slowly, on a PC or Mac.
Finally, Rex Kerr is correct with respect to the "rewarding" of genome enlargment. Genome enlargement was not differentially rewarded; in the particular conditions of the Lenski, et al research, genome length was rendered selectively neutral by awarding SIPs in proportion to genome length. (That's one of my test cases to see if a critic has read the paper sufficiently carefully. Another is the definition of "step" in the paper.)
RBH
*Added in very late edit, but before any responses: There are actually three errors in that first sentence, two explicit and one implicit. The implicit error is one of omission. In the research under discussion, Avida implemented just a subset of "neoDarwinian" evolutionary mechanisms, not the full array. For example, there was no analogue of lateral gene transfer in the configuration used in the Nature paper. An analogue of LGT is available in Avida but was not used in the the Lenski, et al., research. An analogue of sexual recombination is not available in current versions of Avida.
[ 30. January 2004, 08:06: Message edited by: RBH ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 30. January 2004 11:29
Truman made one argument in criticism of the Lenski, et al, paper that has been niggling at me. The argument has been made before (for example, in the earlier thread discussing the Lenski, et al paper in the Literature Forum) but it's particularly clear in Truman's first posting above.
The argument runs thus. From Truman's first posting above: quote:
Avida's organisms replicate thanks to external software and hardware which provide key services; the energy is supplied from external sources (e.g., electricity); their immediate environment is maintained and repaired by humans. Notice that these requirements cannot be damaged by mutations in the Avida model. However, free-living organisms need to have these critical functions encoded on their genes and these CAN be damaged by mutations.
That is, Truman argues that because the experimental context is designed and controlled by humans, one can't generalize to biological organisms "in the wild."
That is precisely equivalent to arguing that, in a bacteriology lab experiment, because the Petri dishes and nutrient media and ambient temperature are controlled by the experimenters and thus "cannot be damaged by mutations," we cannot have any confidence in the results of experiments on the evolution of E. coli performed under those conditions.
I'm aware that intelligent design creationists wish to make a hard distinction between "experimental science" (sometimes called "empirical science") and "historical science," with the implication that the latter is somehow intrinsically less reliable, less dependable, than the former. Pushing that distinction was a goal of the intelligent design creationists in Ohio during the writing of new Science Standards by the State Board of Education last year. They failed in that effort.
But Truman's argument appears to imply that even "experimental science" is an intrinsically flawed and undependable source of knowledge about biological phenomena "in the wild." Because humans control stuff like the energy supply to the Avida world and that world's 'physics' and 'chemistry,' we cannot learn anything interesting about a different world in which those aspects are "naturally" present. Taken to a not-very-extreme conclusion, Truman's critique would imply that neither historical research nor experimental research can provide dependable knowledge, and we would be reduced to passive observation of 'nature in the wild,' hoping for 'experiments of nature' to provide dependable knowledge.
Clearly one must use caution in generalizing wholesale from laboratory experiments to 'nature in the wild.' No researcher I know is unaware of that. Experiments are designed to abstract what are thought to be relevant variables from 'nature in the wild' and to allow manipulating those variables in ways that are potentially theoretically interesting while holding other factors constant. Experiments are not exhaustive simulations of 'nature in the wild.' Nevertheless, experiments can tell us useful stuff about 'nature in the wild,' and Truman's critique must reduce to the observation that one can't generalize wholesale but must be aware of the limitations of a given experiment.
Clearly Lenski, et al, were aware of the limitations of their experiment. Rex Kerr's observation that Truman elided the middle of Lenski, et al's abstract is directly relevant. Much of Truman's critique reduces to a non sequitur, the complaint that Lenski, et al, didn't do something they didn't set out to do!
====================
Incidentally, I mentioned in my posting last night that I had looked briefly at some of the variables Truman complained were biologically unrealistic, particularly the mutation rate. I dug around in some old files and found some relics of those runs. In one run for which I still have the version of the configuration files I used, all mutation rates were reduced to 10% of their default values; SIP 'rewards' were turned off for all but the nine logic functions used by Lenski, et al; SIPs were awarded in proportion to genome length, rendering length selectively neutral; and Avida's analogue of lateral gene transfer, called (disgustingly!) "necrophilia,' was turned on. In that run, which was truncated at 50,000 updates, lineages capable of performing 7 of the 9 logic operations emerged in the population. The exceptions were the two 5-NAND operations (XOR and EQU).
I restarted that same run on an old (and slow) machine some minutes ago, and what Avida calls "species" appear to form more compact spatial clusters on the Avida world plane than occurs under the default conditions. The clusters seem to be 'sharper,' with better-defined borders than those that appear under the default conditions. That is a purely subjective impression, but it's possibly worth looking at systematically. Also, my purely subjective impression is that there is more genotypic variability but less phenotypic variability in the 3,600-critter population under those conditions than in the default conditions, which again may be something worth looking at systematically. In any case, there is clearly quite enough variability to allow lineages capable of performing complex operations to appear in the population.
RBH
[ 30. January 2004, 11:35: Message edited by: RBH ]
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 30. January 2004 13:02
Truman: If he chose to accept the Intelligent Design view of where life came from, he would have some very interesting things to say about his own work! He believes novel binding sites evolved from random sequences. I am convinced most were designed.
Designed by what? Random variation and selection causing the genome to increase its correlation with its selecting environment? The experiment starts with random sequences and ends with correlated sequences mimicking concepts from evolution such as variation and selection.
Truman: In fact, I propose they were originally created with more information content than needed, to provide fault tolerance towards mutations.
An interesting proposal, could you elaborate in some more detail that would help us evaluate this concept?
Truman: Dr. Schneider apparently approved of at least some of the things I suggested on an on-line essay (http://www.trueorigin.org/dawkinfo.asp) in which Dr. Richard Dawkins was on the receiving end, but was not quite so pleased when his own ideas received some of my friendly criticism.
[b]This seems to be a needless ad hominem snipe at Schneider. Schneider 'approved' of the conclusion that information in nature needs an explanation. Schneider's work on Ev seems to provide some of the answers as to a plausible origin of information and complexity in nature. In his response to Truman's 'critique of Ev', Schneider points out various issues and non sequiturs aka strawmen in Truman's response. Schneider argues, and I tend to agree here, that "Many of Truman's complaints have to do with inessential points"
[/b
So let's explore Schneider's arguments about Dawkins and Ev in full context
quote:
I didn't get around to publishing the Ev paper until 2000, 14 years later. What got me going was the creationists who were attacking Dawkins and claiming that it is impossible for biological systems to gain information. (See: Dawkins, R. (December, 1998) The ``Information Challenge''. The Skeptic, 18(4), 21-25 http://www.skeptics.com.au/journal/dawkins1.htm) From the simulation, I knew that was completely wrong, and besides the simulation gives a nice biological result. Dawkins' self-defense was quite weak because he didn't really show how information appears and is measurable in the genome, a job that Rsequence does. The simulation works nicely, showing that starting from random genomic sequences and given only house keeping genes (translation, replication and general metabolism), mutation and selection generate the predicted amount of information in the binding sites. As it turned out, I could also use the simulation to demonstrate that three creationists (Truman, Behe and Spetner) had made serious mistakes. After the publication, in 2001, Dembski attacked my program, thinking he knew where I had "snuck in" information gain. I was able to show that he was wrong on that and I know by the timing and other evidence that he had this information available to him. However, he went ahead anyway and published his error in his book "No Free lunch". This case clearly demonstrates the intellectual poverty of creationist thinking; they cannot admit making mistakes. (See corrigenda.)
History of Ev
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 30. January 2004 20:58
I don't want to drag the discussion off topic, but I would like to comment on RBH's statement that AVIDA does not simulate anything. In fact, I think AVIDA is indeed a simulation, simplified as it is, of a population of imperfect replicators under selection. That this corresponds to certain salient features of biological organisms (or better, their genetic components), which are considered necessary for biological evolution by darwinian mechanims, is what makes AVIDA interesting after all. However, AVIDA is certainly not meant to simulate "cellular networks" (did Truman really mean "molecular networks", by the way? To me, "cellular networks" are the result of interactions between individual cells, such as in the nervous or immune system - rather than interactionds between sub-cellular components, such as the flagellar proteins).
Also, as far as the rate of mutation is concerned, in addition to the point made above that reducing mutation rate would only have a quantitative, not qualitative effect on the outcome, I thought important to keep in mind out that high mutation rates in AVIDA are actually counterbalanced by the low population size within any single run - far below effective population sizes of biological organisms. In other words, most biological populations, due to their size, do in fact sample a dramatic amount of new genetic diversity in each generation, even if their mutation rates (at the individual level) are quite low.
[ 30. January 2004, 20:59: Message edited by: charlie d. ]
IP: Logged
|
|
|
Printer-friendly view of this topic