Adjusting Pre-configuration to Design Outcomes
by BradfordThe Genius Behind the Ingenious is a Biologic Institue article. This appears in it:
Actually, as you may have guessed, attempts to harness the principles of evolution on computers have been underway for many years now. The field dedicated to this undertaking is known as evolutionary computing, and the results are not altogether encouraging for evolutionary biology.
It’s not that evolutionary design has failed on computers—far from it. One of the most celebrated successes, for example, is a NASA antenna that looks like a bent paper clip. [2] It may not be much to look at, but this odd little design works better than any known alternative, which is why NASA has deployed it in space.
A photo of the paper clip can be viewed at the link. The authors explain that the means by which designs are realized entails the assignment of design parameters to virtual prototypes and assessing them with a mathematical model. The idea is to model natural selection by favoring the replication of better prototypes. The better ones would become more prevalent in a virtual population. Mutation is simulated through slight changes in the parameter values of prototypes. Procedural variations to this general approach exist but "culling, replicating, and mutating" persist to derive an acceptable design. This also from the article:
Two major limitations to evolutionary processes seem to assure this. First, it turns out that if you want these processes to go anywhere, you really do need to master the design principles specific to your objective. You’d better believe the NASA team did their homework for the task they were tackling—they knew what materials to use, they knew the range of dimensions to explore, they knew what kind of geometric space to explore, and they knew how to model the performance of any prototype within those specifications. So the software they used was intelligently pre-configured for this particular design task and no other.
This strikes a familiar chord. Pre-configuring software for a design outcome is very much like the concept of pre-configuring genomes to front load them for directed evolution. TT commenter Allen MacNeill recently weighed in on another thread with this comment:
"…natural selection doesn't "produce" anything. As Darwin himself wrote to his friend Charles Lyell, natural selection preserves certain forms and eliminates others. The real "engine" of change in biology is not natural selection, but rather the "engines of variation" that produce the blizzard of new forms, a few of whom survive and reproduce."
Measure engines of variation by the extent of genomic pre-configuration required to enable specified outcomes. The article concludes with these observations:
In the case of antennas, it’s easy to find prototypes that work, and nothing can go horribly wrong if you tweak one of those prototypes. That makes the task of zeroing in on a good design relatively easy. Flash drives, on the other hand, are a different story—no one expects to make a working prototype of one of those by accident, or to improve an engineered version by accident.
The problem, of course, is that the designs of biology look more like flash drives (on steroids) than coat-hanger antennas. Parker definitely got that one right.
So, in light of this, we think it’s time to turn the logic around. If it was reasonable to think that selection ought to be as powerful in virtual worlds as it is in the real world (and we think it was), then the considerable limitations we now see demonstrated (even proven) in virtual worlds should perhaps make us re-think the extravagant claims we make about selection in the real world.
Maybe the intuitions that enable us to recognize brilliant design should be kept in mind when we try to explain it.
November 11th, 2008 at 1:46 pm
Well, they call it "evolutionary computing", but it's really random trial and error. As such, that fact that it works at all, regardless of the difficulties, is proof that random change followed by selection can lead to "specified complexity".
The observation that having the correct variation parameters and selection process allows a telic thinker to believe that the solution itself is "built in", but that's a misdirection. The parameters and process are designed by humans, but that's only because the engineers are trying to achieve specific results.
Yet this is exactly where this approach to engineering differs entirely from evolution in biology: there is no specific goal to evolution, so an improvement that leads to greater reproductive success can be in any direction whatsoever and can be on any scale. The difficulties and limitations the article observes are caused by the human design of the problem, not by the mechanics of the solution.
The article exacerbates this confusion by leaping from antenna design, which is amenable to these techiques because improvement is incremental, to considering flash drives. The comparison of flash drives to biological systems is reasonable, and the idea that these evolutionary techniques would not apply to the problem of designing a flash drive is correct. But this tells us nothing about evolution for the simple fact that the biological systems we observe were not, so far as we can tell, targets identified in advance.
Comment by don provan — November 11, 2008 @ 1:46 pm
November 11th, 2008 at 2:00 pm
But the question is does it lead to observed outcomes and what type of pre-configuaration would be needed?
Comment by Bradford — November 11, 2008 @ 2:00 pm
November 11th, 2008 at 4:18 pm
That's right. It isn't as simple as "specified complexity is impossible."
Would you agree that no pre-configuration is needed if there is no goal? If so it would seem that the question is, "Was there a goal?" The blog entry seems to assume that these complex structures are proof, in themselves, that there was a goal, but what we've found is that they merely raise the question rather than answer it.
To actually answer the question, wouldn't we have to identify examples of pre-configuration?
Comment by don provan — November 11, 2008 @ 4:18 pm
November 11th, 2008 at 4:30 pm
A goal is not my concern. Biological structures are.
Start with a minimally functional genome.
Comment by Bradford — November 11, 2008 @ 4:30 pm
November 11th, 2008 at 8:33 pm
Teleological thinking involves considering the development of biological structures for some predetermined purpose, i.e., a goal. Are you saying you have no telic interest in this topic?
Anyway, the original article was observing problems with using evolutionary programming to achieve specific goals. The point remains that these problems have no bearing on evolution in biology since we are not concerned with goals when considering biological structures.
A minimally functional genome is a preexisting condition. What is your justification for saying it is a pre-configuration?
Comment by don provan — November 11, 2008 @ 8:33 pm
November 11th, 2008 at 11:12 pm
When a course of events is determined in part or in full by initial starting conditions those events were front loaded by initial conditions. A fertilized egg is an example. So too could be series of evolutionary events. Your perception that a biased outcome is not evident is your subjective opinion. There is plenty of data indicating that genes found in "higher species" have their roots in more ancient species.
It has all the prerequisites for cellular life and its diversification set in its (initial) genome and no clear evidence that natural selection would cull a gradual series of changes leading to it.
Comment by Bradford — November 11, 2008 @ 11:12 pm
November 12th, 2008 at 7:03 pm
I'm not sure whether this debate is substantial or merely a disagreement of definitions. Events follow from initial conditions. That's just a definition. Whether those initial conditions are an example of "front loading" would require demonstration of intelligent influence in the intial conditions themselves. (Unless you want to use "front loading" as a synonym for "initial conditions" without any teleological implications.)
This is just assuming your conclusion.
The fact that there's no clear evidence of one path is not evidence that another path involving conscious pre-configuration was taken.
Comment by don provan — November 12, 2008 @ 7:03 pm
November 12th, 2008 at 7:20 pm
It has all the prerequisites for cellular life and its diversification set in its (initial) genome…
It's an observation, not a conclusion.
…and no clear evidence that natural selection would cull a gradual series of changes leading to it.
NS is not a path. It is essential though to OOL pathways. That's why it is presumed to have been a factor in the origin of cells.
Comment by Bradford — November 12, 2008 @ 7:20 pm
November 12th, 2008 at 11:00 pm
Moreover, cells exhibit many design characteristics and processes that human designers use, known telic agents.
Maybe that's not a justification in your mind, but to me it certainly is evidence that warrants a suspicion and more research in that direction. Have you read the Design Matrix?
Comment by kornbelt888 — November 12, 2008 @ 11:00 pm
November 13th, 2008 at 3:34 am
I'm fine with suspicion. If Bradford had said some genome might be a pre-configuration and we should do more research in that direction, I wouldn't have thought twice about it. But he presented it as an identified example of pre-configuration, and that's what I questioned.
Anyway, what research would you suggest that could confirm that some genome at some point in time was a pre-configuration? Could you start by discussing ways of distinguishing a pre-configuration from more mundane initial conditions?
Comment by don provan — November 13, 2008 @ 3:34 am
November 13th, 2008 at 3:47 am
My observation differs from yours. My observation is that some prerequisites were not set in any particular "initial" genome. How can is be that we look at the same thing and make different observations? How do suggest we confirm one observation or the other?
Comment by don provan — November 13, 2008 @ 3:47 am
November 13th, 2008 at 6:25 am
It has all the prerequisites for cellular life and its diversification set in its (initial) genome…
It's an observation, not a conclusion.
Why place quotes around initial? Presumably there was a point in time when no genomes existed and a subsequent point in time when they did. Mention of the term diversification refers to the capacity to adapt which I postulate as a requisite cellular property. Any changes taking place after an initial genome came into existence were made possible by a capacity for replication. Genetic biomolecules are replicating and changing. If you wish to contend that some cellular property, described as a prerequisite for cellular life, came into play after genomes already existed but are not tracable to genetic changes then by all means specify what it is you have in mind.
Comment by Bradford — November 13, 2008 @ 6:25 am
November 13th, 2008 at 5:51 pm
At one point in time there were not genomes. At a later time there are genomes. In between, there's something: currently biology thinks it was some sort of sequence of events. If so, then the decision to mark one specific thing in the sequence "an initial genome" is just an arbitrary dividing line, hence my use of quotation marks. (This actually goes further than this, in that it's not clear when we could legitimately use the term "genome", but I'm just following your lead in using that term for whatever it is we're saying is the hypothetically pre-configured initial condition for the development of life.)
As to me picking a property, that seems like an odd thing for you to ask me to do, since you're the one making a claim. But let's just pick "diversification". An "initial genome" might replicate without any ability to diversify. Then a fluke accident — as opposed to something caused by some quality of the "initial genome" — introduced the ability to replicate with variation. How can you say this definitely didn't happen?
Well, you might say that this is because without diversification, we wouldn't call it a "genome", which is fine, but if you're going to set the minimum requirement for "genome", then you no longer have any justification for calling it an "initial condition".
Comment by don provan — November 13, 2008 @ 5:51 pm
November 13th, 2008 at 7:15 pm
The reverse of what you depict would be much more plausible. A remarkable feature of organisms of all types is the fidelity of their replication. Offspring are very much like their parents and mutations are rare. Selected mutations rarer still. If fidelity was a bit more perfected evolution would not occur. The reason for the great fidelity in copying is also known. Genomes come equipped with self correction mechanisms that are multi-component in nature. The point in mentioning all this is to emphasize that increasing fidelity is accompanied by increasing complexity of error detection and repair functions. They would not exist in a "simple" primordial genome. We would expect such a genome to show a much greater rate of error. This would increase the likelihood of primordial diversification. But an evolutionary process would require a balance between an uncontrolled rate of mutation leading to genomic meltdown and copying perfection which inhibits the capacity for change. This pre-configured genomic balance would form an assumption underlying mutation models.
Comment by Bradford — November 13, 2008 @ 7:15 pm
November 14th, 2008 at 12:57 pm
Once we have minimal fidelity, then evolutionary competition will drive increased fidelity, replication rates, and the exploration of longer sequences.
Comment by Zachriel — November 14, 2008 @ 12:57 pm
November 14th, 2008 at 1:56 pm
So it is said. Postulating what constitutes minimal fidelity and observing newly generated repair mechanisms would do it.
Comment by Bradford — November 14, 2008 @ 1:56 pm
November 14th, 2008 at 2:19 pm
Even relaxed fidelity combined with efficient replication can be advantageous. Catalysis associated with replication has been evolved in vitro with 95% fidelity.
This isn't necessary to the point you raised about " 'simple' primordial genomes". The fidelity in primitive replicators can be very relaxed when counterbalanced against efficiency, and still result in a continuation of the species.
Comment by Zachriel — November 14, 2008 @ 2:19 pm
November 14th, 2008 at 2:50 pm
Simple primordial genomes need to be defined and placed within a selection context. If self-replicating RNA molecules is what you have in mind then fidelity can be defined at one level as the sequence specificity needed for molecular replication. That brings up some additional questions such as how was the sequence specificity derived in the first place and since environmental conditions are critical to and beyond any capacity to regulate via homeostasis like functions, an explanation for conditions is called for. In an ever growing soup of RNA molecules from where are the ribose, bases and phosphates generated. In this scenario the fidelity of their sourcing becomes part of a larger fidelity issue.
Finally, there is the matter of function. The only "function" evident is molecular replication. There is no basis upon which to assume that this scenario results in the generation of systems of interacting parts which replicate as units. For that to occur the self-replicator must become a true genome. That's the context within which fidelity is assessed. If nucleic acid specificity does not lead to systems complexity you have a non-starter.
Comment by Bradford — November 14, 2008 @ 2:50 pm
November 14th, 2008 at 6:02 pm
As long as you agree they are plausible alternatives that contradict the claim, "It has all the prerequisites for cellular life and its diversification set in its (initial) genome", I'll leave it up to you which is more plausible than the other.
That's one possibility.
Comment by don provan — November 14, 2008 @ 6:02 pm
November 14th, 2008 at 6:09 pm
And since we know nothing about primordial genomes, it is impossible to place them in any context. That means we cannot rule out anything on either experimental or logical grounds.
Comment by don provan — November 14, 2008 @ 6:09 pm
November 14th, 2008 at 7:55 pm
Where is the contradiction in the following?
A nucleic acid that is not minimally functional with respect to genomic properties also confers no evolutionary prospects right?
Comment by Bradford — November 14, 2008 @ 7:55 pm
November 14th, 2008 at 9:27 pm
Bradford,
If you are saying that, by definition, a genome must have certain functions, then your definition is what determines the characteristics, not anything about how it came to be such as having been "pre-configured" with those characteristics.
Please keep in mind here that your goal was to show me that you have "observed" this pre-configuration, but so far I'm only seeing definition at work, not observations.
Comment by don provan — November 14, 2008 @ 9:27 pm
November 14th, 2008 at 9:45 pm
I did not invent the term genome. Like other words it has a specific meaning that has nothing to do with "my definition." Why are you having a problem with the term pre-configured? Too telic? Do you prefer more PC verbiage such as "a genome arose"?
Comment by Bradford — November 14, 2008 @ 9:45 pm
November 15th, 2008 at 3:28 am
You invented the term "initial genome" and claimed that it was an identified "pre-configuration".
Comment by don provan — November 15, 2008 @ 3:28 am
November 15th, 2008 at 3:36 am
Indeed. I'd rethink that if someone can point out how life and its evolution are plausble without a genome coming into existence at some point in time. Do you have a link describing the evolution of life in a genomeless environment?
Comment by Bradford — November 15, 2008 @ 3:36 am
November 15th, 2008 at 7:22 am
Bradford:
Notice the shifting goal posts. A self-replicator is not a "true genome".
Comment by Raevmo — November 15, 2008 @ 7:22 am
November 15th, 2008 at 10:45 am
Raevmo:
Right. It's not. You're congusing the chemical identity of a substance which can contain a genetic information with one that actually does.
Comment by Bradford — November 15, 2008 @ 10:45 am
November 15th, 2008 at 11:24 am
Bradford:
I think it's you who is confused. A self-replicating RNA molecule is a genome that contains hereditary genetic information.
Comment by Raevmo — November 15, 2008 @ 11:24 am
November 15th, 2008 at 11:32 am
Then tell me if amino adenosine triacid ester is a genome.
Comment by Bradford — November 15, 2008 @ 11:32 am
November 15th, 2008 at 1:06 pm
Which ape was the first human? Evolutionary processes imply a transition, not a distinct dividing line. This applies to the first human, as well as the first "true" genome. A replicating molecule capable of evolution would usually be considered both genotype and phenotype.
Comment by Zachriel — November 15, 2008 @ 1:06 pm
November 15th, 2008 at 2:07 pm
You did not answer the question Zach. Are you claiming that AATE self-replicated and initiated the pathway that culminated with apes, humans…
Here's ein GedankenExperiment. What would be the progeny of an inorganic compound found to have self-replicating properties? How would heredity guide predictions?
Comment by Bradford — November 15, 2008 @ 2:07 pm
November 15th, 2008 at 2:21 pm
Sure I did. You just don't like the answer. Categorization may not be precise on the edges. These sorts of edges are inherent in many biological situations, e.g. species, population. You are apparently confusing the categorization with the thing itself.
I don't think anyone makes that claim. It just demonstrates the principle that molecules can self-replicate.
Do you mean abiotic compound? The progency of a self-replicator would be more self-replicators—by definition.
Comment by Zachriel — November 15, 2008 @ 2:21 pm
November 15th, 2008 at 2:27 pm
But the claim of raevmo et. al. goes beyond the obvious notation that molecules can self-replicate. He refers to them as genomes. So which self-replicating molecules would be considered genomes? All? None? Or only those one thinks began a chain of events culminating in what we see on earth presently?
Comment by Bradford — November 15, 2008 @ 2:27 pm
November 15th, 2008 at 2:35 pm
A self-replicating molecule is both genotype and phenotype. But again, don't confuse the categorization with the thing itself. If humans evolved, there would have been a time when the distinction between humans and their non-human ancestors was indistinct.
Comment by Zachriel — November 15, 2008 @ 2:35 pm
November 15th, 2008 at 2:38 pm
Zachriel:
But if this is so then we are entitled to an explanation as to the evolutionary pathway of self-replicators like AATE.
Comment by Bradford — November 15, 2008 @ 2:38 pm
November 15th, 2008 at 3:12 pm
Entitled? That's an odd word to use. We would like to have more knowledge concerning the origin of life. That's why researchers propose and test hypotheses.
Comment by Zachriel — November 15, 2008 @ 3:12 pm
November 15th, 2008 at 4:07 pm
Bradford:
I would say all of them. Why exclude unsuccessful ones? Or would unsuccessful genomes be a failure of the Supreme Designer? I guess I don't really see what your problem is with my use of the word genome.
Comment by Raevmo — November 15, 2008 @ 4:07 pm
November 15th, 2008 at 4:39 pm
So this is what the much vaunted scientific position become- faith in unobserved physical processes. Match up ones like the theorized precursor RNA SRMs with an evolutionary pathway and let's pretend others like AATE SRMs replicated themselves to extinction. It's a story without the requisite supporting data.
Comment by Bradford — November 15, 2008 @ 4:39 pm
November 15th, 2008 at 4:48 pm
Bradford:
You're losing it, Bradford. My use of the term genome doesn't imply anything about my take on abiogenesis. The hereditary information in any replicator qualifies as a genome in my book. That's just my abstract definition, and it applies to human genomes, genomes of viruses, genomes of mitochondria, plasmids, SRMs, or even imaginary purely theoretical replicators.
Comment by Raevmo — November 15, 2008 @ 4:48 pm
November 17th, 2008 at 2:28 pm
Bradford,
I will stipulate that there was some genome that could be considered "initial": that is a logical consequence of your definitions. But this does not support your claim that such an "initial genome" is a "pre-configuration".
Comment by don provan — November 17, 2008 @ 2:28 pm
November 17th, 2008 at 2:34 pm
Configured in advance. In advance of what? The advent of the cellular capacity to adapt to their environment.
Comment by Bradford — November 17, 2008 @ 2:34 pm
November 17th, 2008 at 7:32 pm
"Configured" is the part you can't demonstrate. Introducing a new concept — "the advent of the cellular capacity to adapt to their environment" — doesn't help you demonstrate it.
Furthermore, your logical justification for the existence of a genome based on the definition doesn't tell us the initial genome's temporal relation to cells with the ability to adapt. Unless, of course, you are now going to define "adaptation" as "having all the characteristics a genome offers", but that would just more more hand waving along the same lines.
Comment by don provan — November 17, 2008 @ 7:32 pm
November 17th, 2008 at 7:54 pm
You mean I have not introduced a chemical pathway which would configure a genome. I'm not sure such a thing exists. But that should not negate the utility of a computer program intended to harness the principles of change over time.
My working assumption is that the relationship between cells and their genome is defined by our current understanding of cellular biology.
Comment by Bradford — November 17, 2008 @ 7:54 pm
November 19th, 2008 at 3:39 pm
No, I mean you haven't justified using the term "configured". And the rest of your argument depends on us agreeing to use it. It doesn't matter to me how you justify it, just that you do. In fact, I have no clue how one could justify it using "chemical pathways". Chemical pathways don't configure things. Unless, of course, by "pre-configured" you don't mean to imply intelligent action. But if that's the case, I've completely lost your point.
Comment by don provan — November 19, 2008 @ 3:39 pm
November 19th, 2008 at 4:28 pm
Are there specific applications or uses relevant to a genome? Ultimately all cellular functions are affected by virtue of protein synthesis and the regulation of it. So who or what designed, arranged, set-up or shaped an initial genome(s) which replicated and today gives evidence of all descendents? Some unidentified chemical process? God? ET? I don't have to answer that to pose the question requiring a choice of options.
Comment by Bradford — November 19, 2008 @ 4:28 pm
November 19th, 2008 at 8:04 pm
OK. I thought your point was that being "pre-configured" demonstrated an intelligent cause. My mistake.
Comment by don provan — November 19, 2008 @ 8:04 pm