|
Author
|
Topic: Evolving Inventions
|
Janitor@MIT
Member
Member # 125
|
posted 26. February 2003 15:53
Uh, how does expanding a space solve any problem, other than the problem of expanding a space? Some strange ideas here.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 26. February 2003 17:03
A few random comments. First, addressing John, ASCSCommanding wrote quote:
a) Two widely spaced points that could well be two widely spaced points in a single hypervolume. The fin of a lobed fin fish and a horses hoof might suggest inventiveness has taken place, but the intermediaries of the fossil record make that far less clear.
There's a topological presumption in "widely spaced" that requires careful examination. As I've argued a number of times now, an evolving system has several fitness "landscapes," each induced from a single fitness function by different evolutionary operators, and each having different topologies. So, for example, the topology of the fitness landscape induced by point mutation is very different from those of the fitness landscapes induced by duplication mutation or by recombination. Hence "widely spaced" must be defined with reference to the topology of a specific volume induced by a specific operator. A hypervolume of biological possibilities does not have a fixed and static topology. I will have to reread John's essay to remind myself of what he takes to define the topology(ies) of the hypervolumes of which he speaks.
Second, ASCSCommanding remarked that quote:
b) Cases like the flagellum, and other examples you bring up in your earlier post (sexual reproduction, multicellularity, etc.) which are extremely ancient. It seems to me that assessing the inventiveness of a given artifact (be it biological or human) one must have a fairly complete record of what other artifacts existed at the time. How could be determine if a given Babylonian artifact showed TRIZ like inventiveness unless we had a complete record of what other similar artifacts the Babylonians had? The complaint concerning ID is that inventiveness is only found where that kind of record is missing.
On ARN, Alix Nenuphar observed correctly that quote:
The other apparent flaw is how Dembski decides that a given object possesses property X. As Van Till points out, property X (specified complexity) is not really a property of the object, but rather a property of the history of the object. Without knowing the history, property X cannot be assigned.
But Dembski also claims that anything which is designed, has no history (at least of the evolutionary sort that we are concerned with). (emphases original)
That is what makes the probabilities of random assembly (complexity, in Dembski's terms) that are estimated for the bacterial flagellum vulnerable to new learning: The moment a possible/plausible history comes into view the probability associated with chance-and-regularity rises precipitously and the design inference disappears. That is a design-of-the-gaps inference engine. It is also, as ASCSC points out, what makes the attribution of TRIZ-like innovation dependent on outside knowledge of the circumstances surrounding the its appearance and of preceeding and contemporaneous artifacts. Borrowing, for example, can't be ruled out if one doesn't know the larger context from which borrowing might occur.
Third, Micah asked quote:
Isn't Shapiro's basic argument that organisms have the capacity to modify their own genetic code with an adaptational goal "in mind." Are organisms thus capable of exploring outside the original hypervolume? Sorry if this is naive.
As I read him (and I'll be the first to admit I don't yet understand what "natural genetic engineering" really means), Shapiro suggests that over time (or at some time), cells in particular have somehow acquired tools (his word) to tinker with their own genomes. If those tools include the capability to alter the kinds of embedding that John speaks of, then one would expect them to be relevant to John's conjecture. Shapiro explicitly shied away from the question of the origin of those tools in the recent chat and I don't blame him.
Finally, Janitor asked quote:
Uh, how does expanding a space solve any problem, other than the problem of expanding a space? Some strange ideas here.
Expanding a space solves no problems (unless you've just had triplets!). However, as I understand the discussion, the "space" is the space of potential solutions, and enlarging that space increases the likelihood (informal sense) that somewhere in the space there will be solutions. Whether they will be found is a different question.
Much of this is predicated on the erroneous representation of biological evolution as search, and not a little confusion is resulting from that mistaken metaphor. You no doubt have noted that even I, who argues strenuously against representing biological evolution as search, fall into the lingo of search at the drop of a semicolon.
RBH
Edited to add ARN URL.
[ 26. February 2003, 17:18: Message edited by: RBH ]
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 26. February 2003 19:56
John wrote:
quote:
Unless someone brings up a response which shows real insightfulness (in the sense of addressing the evolution of genetic contexts instead of the evolution of individual genes), I'll let this be my last post on this thread. I'm weary of the repititious error in which evolution within a hypervolume (via gene duplication, and subsequent mutation of that gene) is conflated with the re-engineering of the hypervolume itself.
I think the reason why people keep repeating themselves is that it is continues to be quite unclear what you would consider as evidence of a changing hypervolume that could reasonably be expected to exist at this stage of scientific advancement.
You didn't seem to find the 2,4-DNT pathway to be evidence of a changing hypervolume, so apparently creation of new biosynthetic pathways is included in the initial hypervolume.
I'm not sure if you include signalling pathways or not; there are some good examples of recently diverged transcription factors, kinases, and so on. And pretty much everything else is built up from biosynthetic and signalling pathways, so it's not clear what delimits this hypervolume of yours.
What would be an example of hypervolumetric change? You have only been as specific as
quote:
I've listed them in the paper and discussed them earlier on this board: the origin of multicellularity (the origin of many new gene "contexts"), the origin of bodyplans, the origin of sexual reproduction, the origin of eukaryotes, and certainly the origin of life.
This are very hard problems. If you want detailed evolutionary solutions to these, obviously we first have to understand the details of how these systems work! If that is the case, you want the study of biology to be essentially complete. But it's not complete, is it?
So if you want anything you say to be the least bit testable, come up with something more concrete and, well, testable. With the profusion of examples of small changes--including in transcription factors and the like--and in the absence of a clear statement of simple cases of hypervolume change, I have to regard the TRIZ attempt as scientifically irrelevant. Perhaps the argument is a form of refusal to extrapolate; while this can be valid in principle, it is not so useful when we only have enough information to allow extrapolation or nothing.
Perhaps we can revisit the idea in another hundred years or so.
RBH wrote:
quote:
As I've argued a number of times now, an evolving system has several fitness "landscapes," each induced from a single fitness function by different evolutionary operators, and each having different topologies.
Sure, but in reality, each operator has some probability of being applied, and what is relevant is the topology induced by the "I exist" operator, which is some manner of composition of each mutational operator.
So there is no problem with talking about "widely spaced" points; the assumed metric is induced by the "I exist" operator.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 26. February 2003 20:35
Rex wrote quote:
Sure, but in reality, each operator has some probability of being applied, and what is relevant is the topology induced by the "I exist" operator, which is some manner of composition of each mutational operator.
So there is no problem with talking about "widely spaced" points; the assumed metric is induced by the "I exist" operator.
"No problem" is perhaps more than I'd say. I think of the the landscape induced by what you term the "I exist" operator as an ensemble, not a single composite construct. If I think about it a while I may come around to seeing it in the singular. The notion of "neighborhood" gets a little strange for me in that conception, but that is not an insuperable difficulty. In emphasizing the multiple landscapes position I mainly want to break people of (implicitly, unquestioningly) regarding the landscape induced by point mutation as somehow canonical.
I suspect that I think about the "I exist" topology as an ensemble because in some of the work we do we allow the GAs to "choose" to employ or not employ one or another operator. For example, they can evolve to use (or not) what amount to switches to employ (or not) point mutations and within that, the average 'size' of the mutations relative to the quantities representing allele values (we use 16-bit representations of alleles). So that I tend to think of the several operators as independent is an artifact of the way we implement them as independent and GA-selectable operations. Oh, well. Live and learn. ![[Smile]](smile.gif)
RBH
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 27. February 2003 01:16
After having established that Genetic algorithms can indeed increase their hypervolume and thus cannot be in one grand swoop be excluded from being able to generate innovative/creative designs it may be interesting to explore if there may be some examples of such. Such a project however is complicated by the vague definitions of innovative/creative as used in TRIZ and thus on how to recognize creative/innovative solutions from routine design. In order to at least provide some foundation allowing us to define the various forms of design lets discuss the various forms of design.
Gero distinguished between routine and non-routine design. Routine design involves instances in which all necessary knowledge is available or more formally quote:
...that designing activity which occurs when all the knowledge about the variables, objectives expressed in terms of those variables, constraints expressed in terms of those variables and the processes needed to find values for those variables, are all known a priori.
Source: MASS CUSTOMISATION OF CREATIVE DESIGNS John S. Gero
Gero points out that in addition routine design limits the available range of the variables.
Gero identifies two forms of non-routine designing:
Innovative designing and creative designing.
quote:
Innovative designing, in computational terms, can be defined as that designing activity that occurs when the constraints on the available ranges of the values for the variables are relaxed so that unexpected values become possible,
Innovative designing produces designs that belong to the same class as their routine 'brothers' but are also 'new'.
Creative designing:
quote:
in computational terms, can be defined as the designing activity that occurs when one or more new variables is introduced into the design. Processes that carry out this introduction are called “creative designing processes”. Such processes do not guarantee that the artifact is judged to be creative, rather these processes have the potential to aid in the design of creative artifacts. Thus, creative designing, by introducing new variables, has the capacity to produce novel designs and as a result extends or moves the state space of potential designs.
Lets look at the following paper
"Automatic Creation of Human-Competitive Programs and Controllers by Means of Genetic Programming" by Koza et al.
Abstract:
quote:
Genetic programming is an automatic method for creating a computer program or other complex structure to solve a problem. This paper first reviews various instances where genetic programming has previously produced human-competitive results. It then presents new human-competitive results involving the automatic synthesis of the design of both the parameter values i.e., tuning and the topology of controllers for two illustrative problems. Both genetically evolved controllers are better than controllers designed and published by experts in the field of control using the criteria established by the experts. One of these two controllers infringes on a previously issued patent. Other evolved controllers duplicate the functionality of other previously patented controllers. The results in this paper, in conjunction with previous results, reinforce the prediction that genetic programming is on the threshold of routinely producing human-competitive results and that genetic programming can potentially be used as an ‘‘invention machine’’ to produce patentable new inventions.
Koza provides us with two examples in which GA's were used to file innovative design patents
quote:
There are at least two instances where evolutionary computation yielded an invention that was granted a patent, namely a design for a wire antenna created by a genetic algorithm and a patent for the shape of an aircraft wing created by a genetic algorithm with variable-length strings.
Koza continues with a table of 24 examples of "results where genetic programming has produced results that are competitive with the products of human creativity and inventiveness."
15 of these 24 examples involve previously patented inventions, 6 infringe on patents and one improves on a patent. Nine duplicate the functionality of the patent in a novel manner.
The question remains, are these examples of routine or creative/non-routine design?
Koza specifies twoways of running GA's
There are two ways of determining the architecture for a program that is to be evolved using genetic programming.
1 The human user may prespecify the architecture of the overall program as part of the preparatory steps required for launching the run of genetic programming.
2 Architecture-altering operations may be used during the run to automatically create the architecture of the program.
Koza continues on to apply GA to a controller problem in the following manner
quote:
In this paper, programs trees in the initial random generation generation consist only of result-producing branches. Automatically defined functions are introduced sparingly on subsequent generations of the run by means of the architecture-altering operations.
The two lag plant:
quote:
As will be seen below, the result produced by genetic programming differs from a conventional PID controller in that the genetically evolved controller employs a second derivative processing block. As will be seen, the genetically evolved controller is 2.42 times better than the Dorf and Bishop 28 controller as measured by the criterion used by Dorf and Bishop namely, the integral of the time-weighted. absolute error . In addition, the genetically evolved controller has only 56% of the rise time in response to the reference input, has only 32% of the settling time, and is 8.97 times better in terms of suppressing the effects of a step disturbance at the plant input.
The three lag plant:
quote:
As will be seen below, the controller produced by genetic programming is better than 7.2 times as effective as the textbook controller as measured by the integral of the time-weighted absolute error, has only 50% of the rise time in response to the reference input, has only 35% of the settling time, and is 92.7 dB better in terms of suppressing the effects of a step disturbance at the plant input.
In both instances the controller included P, I and D, or proportional constants, integrators and differentiators and the genetic algorithm was allowed to vary its hyperspace by including one or more of each. Not surprisingly the program re-discovers the PID and PI topology as invented by Callender et al.
They conclude quote:
This paper has demonstrated that genetic programming can be used to automatically create both the parameter values tuning and the topology for controllers for illustrative problems involving a two-lag plant and a three-lag plant.
Thus not only did the GA control the parameter values but also the topology allowing the GA to vary the hyperspace.
But not only did the GA find solution but the solutions were better than the best solution provided by experts in the field of control technology.
A propos, Kroo, one of the inventors who patented design in which GA's were used comments that "This configuration was independently "discovered" by a genetic algorithm that was asked to find a wing of fixed lift, span, and height with minimum drag. The system was allowed to build wings of many individual elements with arbitrary dihedral and optimal twist distributions. The figure below depicts front views of the population of candidate designs as the system evolves. On the right, the best individual from a given generation is shown. "
Adrian Thompson describes in "Notes on Design Through Artificial Evolution: Opportunities and Algorithms" an experiment of the design of an electronic circuit in which it was attempted "to allow evolution to explore the design space as a type (c) system, with the minimum or simplifying constraints or prejudice."
A type (c) system refers to a system in which neither the forward nor inverse model is tractable.
quote:
It is expected that the performance of a circuit will fall with rising temperature, but Figure 5 reveals that the evolved circuit's behaviour also degrades as the temperature is decreased from 340mK. This kind of behaviour had never been seen in such proposed `single electron' circuits before, and indicates that the circuit actually exploits or relies upon the thermal noise of the electrons at 340mK. This is not necessarily desirable, and perhaps by evaluating across a range of temperatures during evolution a thermally robust solution could be found [7], but we see immediately that evolution is exploring a previously inaccessible part of design space.
Desirable or not, it is obvious that evolution is exploring new design space.
Finally a paper which I believe I have already mentioned but which captures much of my argument
JOHN GERO AND VLADIMIR KAZAKOV, "ADAPTING EVOLUTIONARY COMPUTING FOR EXPLORATION
IN CREATIVE DESIGNING"
quote:
Abstract. This paper introduces a modification to genetic algorithms which provides computational support to creative designing by adaptively exploring design structure spaces. This modification is based on the re-interpretation of the GA's crossover as a random sampling of interpolations and its replacement with the random sampling of direct phenotype-phenotype interpolation and phenotype-phenotype extrapolation. Examples of the process are presented
And here the relevant part
quote:
Non-routine designing maps onto creative designing. In routine designing all the variables which specify designs are given in advance. This means that the space of possible designs is known a priori, each point in this space can be constructed and evaluated directly. What needs to be done is to search this space in order to locate an appropriate or most appropriate design. The result here is the “best” design from this space. In nonroutine designing the result is the “best” space of possible designs as well as the “best” design from this space. Processes which modify the design space of the search problem are called exploratory processes.
Gero comments
quote:
One of the well-established notions related to creative designing processes is that an important means of characterising them is to determine whether they have the capacity to expand the state space of possible designs - exploration (Gero, 1994).
And finally
quote:
As can be seen from the example the resulting designs are unpredictable in the sense that they are unexpected given only knowledge of the original designs and of the interpolation/extrapolation functions. In this sense the process matches well the meaning of exploration both in the technical sense used in this paper and in the natural language sense. The designs produced by the system demonstrate both the novelty and unexpectedness of what can be generated.
It seems that John was correct in pointing out that creative design requires one to leave the hyperspace of the original and explore different design spaces. As I have shown however, GA's are very capable of doing exactly this, exploring hyperspace by varying the dimensions of the search space. As such I would argue that not only do GA's have the potential for innovative/creative solutions but have actually been shown to exactly produce such designs.
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 01. March 2003 00:02
Ged's long post simply reiterated what I have been writing about throughout this thread in my replies to him. But Ged offers some new comments.
1. Pairwise, there is no loss of funtionality.
In an earlier post, Ged claimed:
quote:
Also that the presence of the new function “already exist in the ancestral genes” is not a pointer to a “simpler ancestor”[EDIT -- I meant to say “more complex ancestor” or simpler descendent], rather it is a pointer to an ancestor of the same complexity.
However, what the paper quite clearly shows is that the result of the gene duplication with regard to Eng1 was not a gene of equal complexity, but of less complexity. There is no evidence that pairwise they assembled as if mimicing the structure and function of Eng1.
Under the classical model, when a gene is duplicated, one of them, mutates very quickly within a short time. However, keeping the eng1 example in mind, the other gene, say eng1a, retains the original function (expressed in both the bud and the hindbrain).
However, the paper shows that rarely (not commonly although Force does point to a paper which shows that this may be common and attempts to explain this possibility through the DDC model), one of the duplicated genes may aquire a new function while the other maintains the original one (where Eng1a is expressed both in the hindbrain and the bud).
The paper next states quite unambiguously that the evidence does not support this view of gene duplication. None of the quotes that Ged offered contradict this, with the exception that Force is very careful to point out that new functions do arise through gene duplication. However, none of these examples pose a threat to TRIZ.
3. The authors take into account the possibility of new functions even within the DDC model.
4. The conclusion of the paper is that greater complexity will be the result of the DDC model in many circumstances, not simplification.
In fact it doesn't. In the conclusion, they quite specifically state:
quote:
If such mutations are dominant and have a negative influence on fitness, then they may be effectively purged from the population and contribute little to the long-term evolutionary dynamics of gene duplicates.
and that:
quote:
empirical and theoretical studies will be required to clarify these issues.
Now although the DDC model does claim that this further facilitates the evolution of new gene functions, what the actual data presented within the paper show does not support any new dimensionality , but less dimensionality. And this data is quite unambiguous.
Gedankin goes on to say:
But why does Alonso not go on to quote from the end of the same paragraph:
Nelson:
The reason is quite simple as the same "possibility" of the facilitation of new function is simply assumed.
quote:
On the other hand, by expanding the time period for which genes are exposed to selection, the preservation of duplicates by the DDC process facilitates subsequent opportunity for the evolution of new functions.
Consider what we thus far know through the empirical evidence:
1. That gene duplication partitions the function of the ancestral gene resulting in no new function but a less complex descendant.
2. The examples of gene duplication we do have are irrelevant to TRIZ.
Here I gave as examples the antifreeze proteins and stated:
The argument here is not one of logical possibility, as we see some duplication events leading to new functions. The problem here is one of capability. If all gene duplication does (commonly, with the expection of a few quite simple examples) is partition the function then we don't really have that much empirical evidence to point to and say that "this is an example of evolutionary inventiveness", when in fact all it is trial and error processes.
As John Bracht discussed:
quote:
quote:
--------------------------------------------------------------------------------
The antifreeze example is even weaker. The lowering of the freezing point of a solute only requires that there be some solvent dissolved in it, so the only requirement for an antifreeze protein is that it be water-soluble. There is not even a need to alter the protein itself, it just has to get accidentally mis-directed into the bloodstream. This would definitely not qualify as an invention in TRIZ; I'm not even sure it would get a patent in the USA. Certainly, no contradiction was overcome and no invention was made.
Now Ged quotes a book which attempts to offer some Biological examples:
quote:
For the Escherichia coli bacterium, a relatively simple organism, it is known that the more than 30% of its proteins are the result of gene duplications (Lazcano and Miller 1994; Riley 1993). Those proteins include its DNA pllymerases, dehydrogenases, ferredoxins, glutamine synthetases, carbomoyl-phosphate synthetases, F-type ATPases, and DNA topoisomerases.
However, these examples further support my points. As Julie Thomas has explained in the past, even taking for granted that sequence simiarlity is a good indicator of gene duplication, what we see with myoglobin and some of the parts of hemoglobin is simply a specialization of function, since they all bind oxygen. Furthermore, take the F-ATP synthase. Some of the subunits in the F-ATP synthase have sequence similarity that may point to gene duplication. However, unless the entire molecular machine arose through gene duplication , then the gene duplication explanation only helps us in determining which parts can be classified under which function, since even assuming a gene duplication event occured, we still have to explain the origin of the four functions of the IC F-ATP synthase, which are the catalytic chamber, the motor, the poston, and the stator. Each part is composed of two subunits.
Moreover, the sequence similarity seen between these subunits can easily be explained from an intelligent design persepective. In other words, the sequence similarity would only reflect the a designer's need to have the same structure conform to the chamber. We don't need to infer gene duplication from an intelligent design perspective.
Now this may seem to be an apologetic in that I am reinterpreting sequence similarity to reflect a common design theme used by the intelligent designer (or designers, with apologies to RBH), where it was interpreted as gene duplication by the authors of Ged's quote. But this can open the door to scientific testing. For example when it comes to two proteins, we can test the two explanations by mutation, where the mutation results in the first sequence becomming the other sequence without losing the tertiary structure.
Thats all I have time for today, I'll reply to Yersinia and RBH some other time.
IP: Logged
|
|
Carl
Member
Member # 677
|
posted 01. March 2003 00:09
It has been some time since the FAQ from Talk Origins about the migration of the jawbones to the earbones was posted.
First, a definition. The species were referred to as intermediate species. Let me suggest that all species are intermediate, except the first and last in a sequence. Rather than try to invest such species with some extra characteristic, I prefer to just call them species.
The proposal being discussed in "Proposed Algorithm for Evolution by ID", another thread here, puts that sequence of changes in an easily explained light. The proposal requires a minimum and maximum amount of change that can be made in any speciation/macroevolution. The steps in that sequence seem to fall within those parameters. The proposal used ID as a driving force. The end result of ID is the human form, and the beginning is the single celled plant. Along the way, good design seems to need the jawbone early in the sequence, but step by step it would be easier to modify an existing part than to institute a new part. Thus each step in the sequence follows logically from the earlier one and leads to the next one. The sequence of changes from Therapod dinosaur to primitive bird is another example of such careful planning.
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 01. March 2003 00:45
Hi Carl
How do you intend to differentiate careful planning from selection/mutation since both would seem to be indistinguishable other than the fact that we do observe mutation and selection.
You mention that humans were the end result but were they the goal? So far there does not seem to be any evidence that humans were the goal of evolution. In fact there are far more succesful species than humans on this world.
In the end however I have to point out that nothing you have said provides any insight into the hypothesis of ID as the driving force? But I would not want to discourage you from proposing in more scientific and testable detail such a proposal.
In Christ
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 01. March 2003 02:26
Alonso said:
quote:
However, what the paper quite clearly shows is that the result of the gene duplication with regard to Eng1 was not a gene of equal complexity, but of less complexity. There is no evidence that pairwise they assembled as if mimicing the structure and function of Eng1.
Of course the papers are available for the reader to read himself/herself:
- Force, A.; Lynch, M.; Pickett, F. B.; Amores, A.; Yan, Y. and Postelethwait, J., 1999 Preservation of duplicate Genes by Complementary, Degenerative Mutations Genetics 151: 1531-1545.
- Lynch, M.; Force, A. 2000 The Probability of Duplicate Gene Preservation by Subfunctionalization Genetics 154: 459-473.
I’ll leave most of the biological detail that Alonso commented on to biologists, and in that regard cite my sources and my original post.
.
First I would comment that Alonso is exactly correct (just as stated in my previous post) that the eng1 becomes involved in less structure, loosing any control of the hindbrain/spinal cord. This is a loss of complexity of what was controlled by the eng1 gene.
What would be more effective is if Alonso had addressed the issue of the complexity of the structures controlled by the en1 gene in the precursor, (rather than the eng1 gene in the decendent,) and then compared the complexity of the structures controlled by the sum or aggregate of eng1 and eng1b as compared to the structures controlled by en1 in the precursor.
As I read the “Preservation” paper:
quote:
Although the expression patterns of engrailed genes are complex, here we focus on expression pattterns of the engrailed-1 gene family in two groups of cells. Zebrafish eng1 is expressed in the pectoral appendage bud, while eng1b is expressed in a specific set of neurons in the hindbrain/spinal cord (Figure 7). …
I see no indication that the Zebrafish, with both a “pecotoral appendage bud” relating to eng1, and a “set of neurons in the hindbrain/spinal cord” relating to eng1b has lost any structure from the precursor. Having both a “pectoral appendage bud” and a “hindbrain/spinal cord” (controlled individually) would seem to be controlling the structures controlled by the original en1, contrary to the appearance of the suggestion. There was no indication that the Zebrafish was simpler in the structures in the sum or aggregate of those regions than any precursor in which en1 became expressed in both regions.
Of course the Force and Lynch et al. papers are discussing the preservation of genes, and not focusing on study of the complexity generated by that preservation (which was left mostly to speculation). Because of that, the particular methods of showing the control aspects may or may not have involved increase/decrease size of structure related to the particular “expression” of a gene -- rather what was of interest was the preservation itself, and the continued control of each gene on a region of the structure. This is precisely the increase in dimensionality (as opposed to some analog of a single axis of control that the single gene represents).
[ 01. March 2003, 02:28: Message edited by: gedanken ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 01. March 2003 14:26
quote:
However, what the paper quite clearly shows is that the result of the gene duplication with regard to Eng1 was not a gene of equal complexity, but of less complexity. There is no evidence that pairwise they assembled as if mimicing the structure and function of Eng1.
Ged:
Of course the papers are available for the reader to read himself/herself:
Force, A.; Lynch, M.; Pickett, F. B.; Amores, A.; Yan, Y. and Postelethwait, J., 1999 Preservation of duplicate Genes by Complementary, Degenerative Mutations Genetics 151: 1531-1545.
Lynch, M.; Force, A. 2000 The Probability of Duplicate Gene Preservation by Subfunctionalization Genetics 154: 459-473.
I’ll leave most of the biological detail that Alonso commented on to biologists, and in that regard cite my sources and my original post.
Nelson:
Yes I encourage the reader to obtain the two papers and focus on the actual empirical data rather than the assumptions concerning the capabilities of gene duplication.
Ged:
What would be more effective is if Alonso had addressed the issue of the complexity of the structures controlled by the en1 gene in the precursor, (rather than the eng1 gene in the decendent,) and then compared the complexity of the structures controlled by the sum or aggregate of eng1 and eng1b as compared to the structures controlled by en1 in the precursor.
Nelson:
Gene eng1 is expressed in the bud, eng1b is expressed in the hindbrain. But when they look at the unduplicated gene (En1) in mice and chicken, it is expressed in both the buds and hindbrain ganglia. Now, it seems to me that functionality is a good indication of what kind of complexity is being caused by the gene duplication. Eng1b cannot be expressed in bud and eng1 cannot be expressed in the hindbrain. However, En1 can be expressed in both. This is not an increase in dimensionality, this is a decrease. Ged talks about structure but as far as structure is concerned, we are not talking about the size or the thing, we are talking about the functional space in which evolution can work with.
Here are some excerpts from Bracht's post that I think are relevant:
quote:
Each Hox gene has a set of enhancers that ensure that it gets expressed only in the correct stripe down the embryo. This set of enhancers is set up such that only in the target tissue does just the right combination of transcription factors activate it (these transcription factors are set up in gradients in the embryo). This actually makes it difficult to evolve a "new" stripe, because you can't just duplicate the gene. You have to evolve new enhancer elements which will bind to the unique combination of transcription factors (both enhancers and repressors) that exist in the target tissue.
and
quote:
you want to produce a new structure, like a new appendage, it's not enough to duplicate a gene, or even a bunch of genes. It (they) will simply be wired into the original position, in the old gene network. You've got to add new regulatory elements that cause the appendage to be turned "on" in the right tissues. You also have to have the right downstream differentiation gene batteries that form different parts of the appendage and add muscles, nerves, etc. You've got to add all this genetic programming to get the appendage in the first place--before it can evolve. All this genetic machinery is necessary to properly "interpret" the duplicated gene in a new way. Sure, once it's there, the new appendage can evolve non-inventively to get longer, or shorter, or whatever. But the inventive change is re-wiring the genetic regulatory system.
Back to whether simply a duplicated gene has the promise of becomming something more than it's "less of a man" state, well thats just begging the question I'm afraid.
[ 01. March 2003, 16:35: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Frances
Member
Member # 169
|
posted 01. March 2003 14:31
Nelson,
Could you let us know if at least you agree that in principle gene duplication provides the genome with a change hypervolume? We may explore if evolutionary algorithms can actually use this to their advantage bit at least in principle GA's are not confined to their initial hypervolume.
Or do you disagree?
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 01. March 2003 15:59
Yersinia:
Since when did "ICness" have anything to do with method of assembly? Are you telling me that a mousetrap, identical to Behe's mousetrap, that was not "built from scratch" (whatever that means, exactly -- it's not like earbones are carved out of a leg bone or something), it would magically cease to be IC?
Nelson:
No that has absolutely nothing to do with what I wrote. I never said that something ceases to be IC just because it is being viewed at the morphological level. Anything can be applied to the definition of IC, a whole organism, eco-systems, the entire universe can be viewed as IC with enough meandering around the point of IC, which is to be relevant to evolutionary pathways.
ID proponents and Darwinian evolutionists are interested in the origin of IC systems. This has nothing to do with whether something can be called IC. So when you say that a wing is IC or the ear-bone system is IC, and you say "the origin of this IC system can be understood through evolution" I must question what relevance this has to the origin of molecular IC whose assembly is completely different from how morphological features are developed whether thing truly evolved or not.
Nelson:
Furthermore, it's another quite simple example of the reshuffling of pre-existing parts, and 3 measly parts at that. So even if this was a molecular example, it would be just as trivial as the PCP degradation pathway or the antifreeze example.
Yer:
What I find *really* entertaining is how, with practically every example I raise, the reaction is something like (1) Much doubt and objections that there is no evidence that the example in question evolved
Nelson:
I never raised any objection about whether this thing evolved. I am simply taking it for granted that this fossil evidence is consistent with an evolutionary interpretation.
Yer:
(2) followed by argument and citation of more evidence followed by
Nelson:
Well, there is no good evidence that these things evolved through RM&NS. However, again, thats not important at the moment. My interest in IC (and ID) has always been that it's focus is on the stuff of evolution, what evolution is built on.
Yer:
(3) an assertion that, OK, maybe the example evolved but it's really trivial and doesn't meet whatever ill-defined IDist criteria is popular this week.
Nelson:
This is not an ill-defined criteria. We are dealing with molecular systems that are composed of hundreds of parts, whereas this is just a 3 part system that changed due to regulartory and proteins any one of which cannot be removed without losing function. Heck I'll grant that they are IC, I'll even grant that they evolved. But these changes are extremely dependant on regulatory and expression changes. Why would you ignore this?
Yer:
...set of finely-crafted, delicately matched, sound-conduction devices -- from jawbones, of all things -- a "trivial" acheivement.
Nelson:
Suddenly an example of Biology is finely-crafted and delicately matched while terribly complex molecular systems with cooling systems, a rotary torque accomplished which dazzles modern engineers, and finely tuned assembly is inefficient and easily evolvable. I don't get it.
Yer:
Despite the fact that, as the above graphic shows, it represents the very defining characteristic of mammals that distinguishes them from mammal-like reptiles.
Nelson:
Whether a feature defines a characteristic of a certain lineage of animals is irrelevant to my point about the origin of IC systems. This is an example of three bones in the the middle ear, while reptiles have only one. If you put these fossils in a chronological sequence of therapsids, you can see two of the small reptilian bones near the attachment point of the jaw get smaller and gradually become incorporated into the middle ear. How the heck is this relevant to the 20-part IC system, even forgetting about the distinction between the molecular level and the morphological level?
But that is an important distinction. Consider the fact that these morphological changes are due to regulatory changes which is why the molecular machinery of life has stayed more or less the same even in morphologically different organisms.
Yersinia:
Humans have even given the bones names on analogies with human devices, and refer to their function in terms of lever action, impedance matching, etc. The analogy with a mousetrap is rather close -- transfer of energy via physical mechanisms such as levers, to deliver it somewhere else. If you concede that such a thing is a trivial thing for evolution to accomplish, then you've just conceeded that IC is a trivial thing for evolution to produce, and there goes your argument for ID.
Nelson:
Again, the bone system is no comparison whatsoever to molecular machines like the bacterial flagellum, or even the 5-part mousetrap, which must be assembled without the help of a developmental program. You are completely ignoring how evolution of morphological features are built on the evolution of development, and molecular systems are not. I'm not saying that someone can't give a thorough ID analysis taking IC into account when it comes to morphological features, in fact, I have a feeling that is currently being worked on or willb e worked on in the future. But with any biological feature given any label, IC or whatever, we have to work with the context of what we are dealing with.
Yer:
Regarding gene duplication and the origin of information, you've got a lot more to deal with than just antifreeze genes. There are dozens of published examples of the detailed origin of new genes, although since no one here has given a non-goalpost-moving definition of "inventive" or "technical contradiction" there is no way to know which ones would meet the criteria.
Nelson:
Please give me an example of how we have moved the goal post with regard to TRIZ. All you have given us are trivial examples that exemplify exactly the trial and error processes that the paper points to. How in the world is a sluggish PCP degradation pathway or an inevitable consequence of a trivial gene duplication an example of a technological invention such as the F-ATP synthase?
Nelson:
Although the type III systems are simpler than flagella, they are also IC, and not only that, to tell a co-option story about how you can add the other parts of the flagellum, requires co-option of multi-component systems. Again, see the 6-part essay at ]http://idthink.net
Yersinia:
Are you telling me that all those parts of a Type III secretion system are really required for the function of exporting proteins? All of those simpler protein export systems must not be functional then...
Nelson:
What are you talking about? I can catch a mouse with a simpler 2-part mousetrap. So what? The Type III machinery is IC.
Yersinia:
As for co-option of multi-component systems, what's so hard about that? That's exactly what happened in the "trivial" case of mammalian middle-ear bones. "Multi-component cooption" is really just the linking of two systems. With earbones it happened as gradually as can be.
Nelson:
Whats hard about it is the shier number of interacting parts involved and their subfunctinos (and mini-IC systems). A co-option tale with the 3-bone system is quite simple to imagine even forgetting my objections concerning development above. It's the same material that can be reduced to one part (or one bone). A 3-part system evolving through co-option is a lot different than a 50 part system evolving through co-option. It's sort of like having your air conditioner start working with your engine, as Orr put it (or something like that).
Yer:
You may have been implying another version of the eternal "all the parts came together at once" strawman version of cooption, but no biologists advocate that model.
Nelson:
No where throughout my post do I even mention such a scenario.
Yersinia:
As for molecular examples, please explain how the following is not overcoming a technical contradiction.
1) In the beginning, bacteria could not degrade 2,4-DNT.
2) Humans introduced 2,4-DNT, a synthetic chemical, into the environment
3) Bacteria developed, in quite a complex manner, a way to degrade it. And just in a few years.
Nelson:
I have to read the paper in order to fully comment on it which I will do in the near future. But something tells me it's as irrelevant as the PCP example.
[ 01. March 2003, 16:39: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 01. March 2003 16:33
Rex:
Nelson, when you have a brood size of about 200, 2-3% survivors is plenty for natural selection to act upon, if the change really does provide some major selective advantage--for example, allowing the organism to exploit a previously empty niche. Also, the worms are self-fertilizing hermaphrodites, and as such, under laboratory conditions, are genetically identical. The 2-3% survivors presumably survive for stochastic and/or environmental reasons.
Nelson:
I don't understand what the above has to do with my post, as I showed quite clearly that the change that occured was irrelevant to evolution in that the 2-3% survived because they were symmetrical and not despite they were asymmetrical, (and I wrote the reasons why). From an evolutionary perspective, the only way I can interpret these results is that they are a relic of the last ice age, the ones that developed their axis correctly in low temperatures had a selective advantage and reproduced more offspring than the ones that had "normal" problems directing their developmental axis. But thats about it.
Rex:
With regard to the flagellum, I will simply reiterate my earlier point that the T3SS is a functional complex that indicates that there could be indirect functional intermediates between zilch and a flagellum, and that IC does not rule out evolution by indirect functional intermediates. Behe, Darwin's Black Box, p. 40:
Nelson:
Indeed, but as I pointed out in my original post, these pathways involve nothing but raw chance as each multi-component system just happens to fortuitously interact with another mutli-component system and all the way up to the flagellum, whose function is not realized until all these mutli-component parts are in place. That sounds more like teleology than a stochastic process to me.
Rex:
In the case of the flagellum, that precipitous drop looks much less precipitous in the presence of an example of a complicated near-subset with a different function.
Nelson:
It is exactly because it is not a near subset that the pathway becomes suspect. The export machine is addded to a 9 part filament, and then invokes another co-option of a 3 part machine. Not to mention that the subset you are referring to post-dates the flagellum. No, it does not look less precipitous at all.
To the moderator:
By the way I am dying to get involved in the other threads on this board (especially the Shapiro one), but I noticed that I am responding to three people here which takes up my three posts per day limit. I hope the moderator doesn't mind me responding to next replies in the future to my three posts here in one post as I seen it done before by other posters. Although I know that the consequence of this would be that I would have to keep my replies brief.
[ 01. March 2003, 16:36: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Moderator
Administrator
Member # 1
|
posted 01. March 2003 18:46
The three post limit is only intended as a guide. Since your posts are substantial and "on point" and contribute to the overall health of this board, do not fret too much about "exceeding the 3 post limit."
IP: Logged
|
|
gedanken
Member
Member # 594
|
posted 01. March 2003 19:18
quote:
Ged:
What would be more effective is if Alonso had addressed the issue of the complexity of the structures controlled by the en1 gene in the precursor, (rather than the eng1 gene in the decendent,) and then compared the complexity of the structures controlled by the sum or aggregate of eng1 and eng1b as compared to the structures controlled by en1 in the precursor.
Nelson:
Gene eng1 is expressed in the bud, eng1b is expressed in the hindbrain. But when they look at the unduplicated gene (En1) in mice and chicken, it is expressed in both the buds and hindbrain ganglia. Now, it seems to me that functionality is a good indication of what kind of complexity is being caused by the gene duplication. Eng1b cannot be expressed in bud and eng1 cannot be expressed in the hindbrain. However, En1 can be expressed in both. This is not an increase in dimensionality, this is a decrease. Ged talks about structure but as far as structure is concerned, we are not talking about the size or the thing, we are talking about the functional space in which evolution can work with.
Thanks, Alonso, for not claiming that there was a reduction in complexity of the overall organism between the organism with En1 and the organism with (eng1+eng1b).
My point was not to demonstrate an increase in complexity with this, but to demonstrate the essential separation of control. As I have repeated, well repeatedly, presence of a new dimension does not imply movement in that new dimension.
I agree that there is not an increase in complexity, nor a decrease in complexity demonstrated here of the overall organism, with regard to the before or after duplication events. (And once again repeated). So this is not to be taken as an indication that I think there was an actual increase of complexity demonstrated in this case. And in fact just as Alonso has repeated again there was a decrease in complexity of what was being caused by the individual gene (each duplicated gene), but without a decrease in complexity of the overall organism with the combination of genes.
I have a much longer presentation planned, but I am already over my 3 post limit I believe, so I shall leave this for a later day (and time to finish it).
For now consider this. The En1 has the capacity to relate to construction of both buds and hindbrain. This means that En1 shows the potential of this degree of complexity of what it causes because of the actuality of its expression demonstrates precisely that. Now the eng1 gene is controlling aspects of the hindbrain only (not the “bud” as Alonso has correctly identified and now is doing less complex control). But eng1 gene is of the same size as en1. What is the potential for the degree of complexity of the structure that is possible? What if the degree of complexity it controlled were to increase back to that of En1 in a future mutation? We already have information that a gene of the size eng1 (about the same size as en1, since it was a modified version of precisely en1) can produce complexity of both hindbrain and “bud”. So there is evidence that a modified version of eng1 could produce equally greater complexity in some descended modification. Since it exists alongside of eng1b, the total complexity produced could at that point be greater than was possible with simple modification of en1 standing alone. But now that new complexity, if it should happen as a result of a mutation of the eng1, can go somewhere else, and is not limited to simultaneously affecting both hindbrain and bud in a combinational way.
The essence is not a sudden large increase in dimensionality. Rather there is only the smallest change being affected here, and any really noticeable change will probably require a significantly larger number of changes. (I leave that to biologists, whom I am not among.) But it seems to me to be very significant that Force et al. have demonstrated that the eng1 and eng1b are duplicated, still have control of something, and are staying around. Since they have the potential of returning to the level of complexity of effect of en1 for each gene, there is potential demonstrated here for future complexity.
And then look at the difference between the ancestors and the descendents that have been analyzed to gain the very information of lineage that is assumed in this discussion. If the lineage is correct, then there is obvious differences in what the genetic information of the different genomes can create. There are obvious differences of structure. Surely these differences, observed as dependent on the very differences of genome that we are observing, have a relationship to the complexity generated by those genomes -- and that there is a possible pathway with descent with modification operators that are implied by the very analysis that gives evidence they had a lineage in the first place.
[ 01. March 2003, 19:47: Message edited by: gedanken ]
IP: Logged
|
|
|
Printer-friendly view of this topic
