Get Out of Jail Free: Playing Games in an RNA World
Evolution News & Views September 23, 2013 5:20 AM
Four Darwinian mathematicians and biologists from New York University (one from Puerto Rico) think that RNA molecules played games to invent life. Even if the RNA could spontaneously form, why would mindless molecules scheme to create a universal, nearly optimal genetic code via a pointless game?
Jee, Sundstrom, Massey and Mishra, writing in the Royal Society Interface, ask, "What can information-asymmetric games tell us about the context of Crick's 'frozen accident'?" Francis Crick viewed the origin of the genetic code as an accident that caught on and became universal. But how did gene sequences become associated with polypeptide sequences having function? They know that the genetic code, as is, is pretty darn good:
The genetic code, the mapping of nucleic acid codons to amino acids via a set of tRNA and aminoacylation machinery, is near-universal and near-immutable. In addition, the code is also near-optimal in terms of error minimization, i.e. tRNAs recognizing similar codons may be mistaken for each other during translation, yet these mistakes often have no negative impact on translation because similar codons map to identical amino acids or ones with similar physiochemical properties. Biochemists have long wondered: If immutability and universality were early properties (i.e. the genetic code was a "frozen accident"), then how could natural selection encourage error-minimization? If selection for an error minimizing genetic code predated immutability and universality, then why is the standard code less than optimal? (Emphasis added.)
Although "numerous models have been proposed" to explain this "apparent paradox," they each have problems, such as "premature freezing" of the code, or in the case of neutral evolution, inability to explain the code's universality. So these guys enter the fray.
Like Crick, they know that hitting upon a functional enzyme by chance in the space of random polypeptides is improbable to the extreme:
Because of the relative length and complexity of modern enzymes, it may be possible that the earliest peptides were not enzymes in the traditional sense. To "accidentally" stumble upon genes encoding such enzymes at the same time an error minimizing code occurred by chance, as suggested by Crick, has vanishingly small probability.
Their job, therefore, is to find pointless polypeptides associating with pointless polynucleotides in some sort of "signaling game" that makes them both "help" each other over time until universality, immutability and optimality reach an equilibrium that just happens to be near maximum. Their very helpful tool in this endeavor is game theory:
As suggested by Maynard-Smith, games in a biological setting, unlike traditional ones in game theory, might not require "rational agents." A population of animals of the same species, for instance, may over the course of evolution behave according to game-theoretic principles even though none of those animals is a "rational agent," in a traditional sense. A species may "learn" over evolutionary time to select certain behaviors through random mutations, genetic drift, and selection, and ultimately reach a Nash equilibrium, in this case defined as an evolutionarily stable state in which each agent does not deviate strategies so long as all other agents in the system also do not deviate from their adopted strategies. "Utility" in the game-theoretic sense physically manifests as reproductive fitness.
They put "utility" in quotes, because it takes a rational agent to determine what is useful. What they are looking for is an equilibrium between mindless players aiming nowhere. Life and optimal coding become incidental byproducts of the equilibrium. Is there any other chemical reaction in nature that arrives at such coding specificity without trying? One might get an oscillation between states, but not a code that specifies a function.
Overall this paper presents a framework for studying signaling game dynamics in instances where both message length and distortion are factors in the utility of both senders and receivers. Although we have applied the framework here primarily to the evolution of the genetic code, similar analyses might be applied to the evolution of many other seemingly fixed processes, where the evolutionary clock appears to have frozen a biological process prematurely to an arbitrary conventional structure.
Well, best of luck. We find them personifying the molecules. The molecules adopt "strategies." They "learn" over evolutionary time. They send "information" or receive it, as they "signal" each other with "messages." Does this make any sense? Take out the words implying personality, goal and purpose, and the idea seems silly, much more so than for antelope strategizing to outwit a lion. These are just dumb molecules!
It's not necessary to delve into the equations of their "game," because math cannot rescue a bad premise. What we find them doing is weaving a fantastic tale in their own imaginations, starting with already-existing complex molecules in a mythical RNA world (which has its own problems).
It is usually hypothesized that the genetic code formed in the context of an RNA world, gradually exposed to an emerging amino acid world. We envision a scenario with two agents: proto-mRNA (strings of codons with information) and sets of proto-tRNA (RNAs with distinct anticodons, each able to bind a particular amino acid). In a given generation proto-mRNA and a particular set of proto-tRNA interact. The pair replicates via RNA replicase ribozymes. However, they may also chemically aid their own replication through the accurate production of proteins (possible identities of these proteins are stipulated in Discussion).
These gamers assume the existence of (1) RNA ribozymes capable of replication, (2) information, (3) transfer RNA with distinct anticodons, (4) accurate production of proteins. Who, we might ask, "usually hypothesized" such things? They should be dismissed from the science lab on account of "envisioning scenarios" instead of doing real chemistry.
Many other problems are completely ignored or glossed over in their visionary scenario, such as the problem of getting one-handed amino acids and sugars by chance. They also assume that natural selection would operate at the scale of molecules in an RNA world before life -- a fallacy, because natural selection requires not just replication, but accurate replication, accurate enough to avoid error catastrophe.
The news release from New York University, as expected, sanctifies this proposal as the inspired work of genius professors. It also won the uncritical acclaim of Science Daily and other news outlets: "Researchers have created a model that may explain the complexities of the origins of life." Be sure to thank the NSF for funding this paper in a down economy.
Well, It Could Happen
Throughout this weird paper, the authors display reckless imagination with frequent assertions that various miracles of chance "could" or "may" or "might" happen. (If a pig had wings, we all know, it "could" fly, provided it also had flight muscles, feathers, avian lungs, and all -- watch Flight.) Added to the heavy spicing of "possibility" words, they frequently endowed the molecules with goal-directed behavior, personifying them as willing game players. Here is but one egregious example from the abstract:
Such a framework suggests that cellularity may have emerged to encourage coordination between RNA species and sheds light on other aspects of RNA world biochemistry yet to be fully understood.
So, out of nowhere, "cellularity emerges" to "encourage coordination." Are you seeing any light that has been shed yet? Later, the personification, assumed goal-seeking, and speculation gets even worse:
The model presented here demonstrates that the modern genetic code evolved most likely by a combination of previously hypothesized forces, involving neutral and selective evolution. Whereas a natural predisposition toward an error-minimizing code is not a necessary condition for an optimized genetic code, neutral evolution may have been an important force in establishing universality. At the same time, selective pressure can provide a powerful impetus for a genetic code to move toward error-minimization and, somewhat surprisingly, also enforce its immutability so as to maintain compatibility with the genome.
Who does the enforcing? Who does the establishing? Who does the maintaining? Who follows an impetus to move toward error minimization? What is an error, anyway, to a mindless molecule? This is crazy, but not crazy enough for the Royal Society to publish it.
They get away with this because it fits the requirement of naturalism: "No intelligence allowed." Within that constraint, they follow Finagle's 6th Rule: "Do not believe in miracles. Rely on them."
Good-bye, RNA World
The authors feel somewhat justified in "envisioning" their make-believe "scenario" on the grounds that "Evidence for such a world [RNA world]... is growing." Too bad this paper came out about the same time that Steven Benner, a veteran origin-of-life researcher, poured cold water on the idea at the Goldschmidt Conference in Florence in August. Here's what he said happens to ribose (an essential sugar for RNA) and other biomolecules when exposed to the watery conditions assumed on the early earth, according to an NBC News article:
The early environment on Earth, however, was challenging to the rise of life as we know it, at least in Benner's view. One of the biggest challenges has to do with the process by which organic molecules gave rise to life's chemical building blocks: RNA, DNA and proteins.
If left to themselves, adding energy to organic molecules just tends to turn them into tar or an oily substance. That's what Benner calls the "tar paradox": How could organic materials ever give rise to biopolymers like DNA?
Science Magazine describes the depressing picture:
However and wherever life began, one thing is sure: Its first organic building blocks, called hydrocarbons, had a number of hurdles to clear before evolving into living cells. Fed with heat or light and left to themselves, hydrocarbons tend to turn into useless tarlike substances. And even when complex molecules like RNA (most biologists' best guess for the first genetic molecule) arise, water quickly breaks them down again.
The RNA-world scenario is so hopeless, in fact, that Benner took the extreme step of claiming that life must have formed on Mars (on dry ground under special conditions), and then got transported to earth via meteors. While some reporters leaped onto the sci-fi suggestion that "We may all be Martians!" (e.g., Space.com), thinking people will surely catch the cry of desperation in such a proposal.
Conclusions
So, even if one were willing to grant the time of day to Jee et al.'s "game theory" notion, Darwinians can't even get the starting materials to play with. It would be more realistic for them to start with balls of tar, and racemic biological gunk broken down by water.
Any way you slice it, the "game theory" approach of these imagineers is an exercise in futility. And that's before even thinking rationally about the problem of the origin of genetic information, discussed in depth in Stephen Meyer's Signature in the Cell.
What a crazy world Darwinism and methodological naturalism (MN) has bequeathed us. The way out is to relax the arbitrary MN rule, to think outside the naturalistic box, and once again, to follow the evidence where it leads. Optimized codes do not "arise" from "frozen accidents." From our universal experience, they are products of intelligent design. That's no game. That's no "scenario." It's reality.
Evolution News & Views September 23, 2013 5:20 AM
Four Darwinian mathematicians and biologists from New York University (one from Puerto Rico) think that RNA molecules played games to invent life. Even if the RNA could spontaneously form, why would mindless molecules scheme to create a universal, nearly optimal genetic code via a pointless game?
Jee, Sundstrom, Massey and Mishra, writing in the Royal Society Interface, ask, "What can information-asymmetric games tell us about the context of Crick's 'frozen accident'?" Francis Crick viewed the origin of the genetic code as an accident that caught on and became universal. But how did gene sequences become associated with polypeptide sequences having function? They know that the genetic code, as is, is pretty darn good:
The genetic code, the mapping of nucleic acid codons to amino acids via a set of tRNA and aminoacylation machinery, is near-universal and near-immutable. In addition, the code is also near-optimal in terms of error minimization, i.e. tRNAs recognizing similar codons may be mistaken for each other during translation, yet these mistakes often have no negative impact on translation because similar codons map to identical amino acids or ones with similar physiochemical properties. Biochemists have long wondered: If immutability and universality were early properties (i.e. the genetic code was a "frozen accident"), then how could natural selection encourage error-minimization? If selection for an error minimizing genetic code predated immutability and universality, then why is the standard code less than optimal? (Emphasis added.)
Although "numerous models have been proposed" to explain this "apparent paradox," they each have problems, such as "premature freezing" of the code, or in the case of neutral evolution, inability to explain the code's universality. So these guys enter the fray.
Like Crick, they know that hitting upon a functional enzyme by chance in the space of random polypeptides is improbable to the extreme:
Because of the relative length and complexity of modern enzymes, it may be possible that the earliest peptides were not enzymes in the traditional sense. To "accidentally" stumble upon genes encoding such enzymes at the same time an error minimizing code occurred by chance, as suggested by Crick, has vanishingly small probability.
Their job, therefore, is to find pointless polypeptides associating with pointless polynucleotides in some sort of "signaling game" that makes them both "help" each other over time until universality, immutability and optimality reach an equilibrium that just happens to be near maximum. Their very helpful tool in this endeavor is game theory:
As suggested by Maynard-Smith, games in a biological setting, unlike traditional ones in game theory, might not require "rational agents." A population of animals of the same species, for instance, may over the course of evolution behave according to game-theoretic principles even though none of those animals is a "rational agent," in a traditional sense. A species may "learn" over evolutionary time to select certain behaviors through random mutations, genetic drift, and selection, and ultimately reach a Nash equilibrium, in this case defined as an evolutionarily stable state in which each agent does not deviate strategies so long as all other agents in the system also do not deviate from their adopted strategies. "Utility" in the game-theoretic sense physically manifests as reproductive fitness.
They put "utility" in quotes, because it takes a rational agent to determine what is useful. What they are looking for is an equilibrium between mindless players aiming nowhere. Life and optimal coding become incidental byproducts of the equilibrium. Is there any other chemical reaction in nature that arrives at such coding specificity without trying? One might get an oscillation between states, but not a code that specifies a function.
Overall this paper presents a framework for studying signaling game dynamics in instances where both message length and distortion are factors in the utility of both senders and receivers. Although we have applied the framework here primarily to the evolution of the genetic code, similar analyses might be applied to the evolution of many other seemingly fixed processes, where the evolutionary clock appears to have frozen a biological process prematurely to an arbitrary conventional structure.
Well, best of luck. We find them personifying the molecules. The molecules adopt "strategies." They "learn" over evolutionary time. They send "information" or receive it, as they "signal" each other with "messages." Does this make any sense? Take out the words implying personality, goal and purpose, and the idea seems silly, much more so than for antelope strategizing to outwit a lion. These are just dumb molecules!
It's not necessary to delve into the equations of their "game," because math cannot rescue a bad premise. What we find them doing is weaving a fantastic tale in their own imaginations, starting with already-existing complex molecules in a mythical RNA world (which has its own problems).
It is usually hypothesized that the genetic code formed in the context of an RNA world, gradually exposed to an emerging amino acid world. We envision a scenario with two agents: proto-mRNA (strings of codons with information) and sets of proto-tRNA (RNAs with distinct anticodons, each able to bind a particular amino acid). In a given generation proto-mRNA and a particular set of proto-tRNA interact. The pair replicates via RNA replicase ribozymes. However, they may also chemically aid their own replication through the accurate production of proteins (possible identities of these proteins are stipulated in Discussion).
These gamers assume the existence of (1) RNA ribozymes capable of replication, (2) information, (3) transfer RNA with distinct anticodons, (4) accurate production of proteins. Who, we might ask, "usually hypothesized" such things? They should be dismissed from the science lab on account of "envisioning scenarios" instead of doing real chemistry.
Many other problems are completely ignored or glossed over in their visionary scenario, such as the problem of getting one-handed amino acids and sugars by chance. They also assume that natural selection would operate at the scale of molecules in an RNA world before life -- a fallacy, because natural selection requires not just replication, but accurate replication, accurate enough to avoid error catastrophe.
The news release from New York University, as expected, sanctifies this proposal as the inspired work of genius professors. It also won the uncritical acclaim of Science Daily and other news outlets: "Researchers have created a model that may explain the complexities of the origins of life." Be sure to thank the NSF for funding this paper in a down economy.
Well, It Could Happen
Throughout this weird paper, the authors display reckless imagination with frequent assertions that various miracles of chance "could" or "may" or "might" happen. (If a pig had wings, we all know, it "could" fly, provided it also had flight muscles, feathers, avian lungs, and all -- watch Flight.) Added to the heavy spicing of "possibility" words, they frequently endowed the molecules with goal-directed behavior, personifying them as willing game players. Here is but one egregious example from the abstract:
Such a framework suggests that cellularity may have emerged to encourage coordination between RNA species and sheds light on other aspects of RNA world biochemistry yet to be fully understood.
So, out of nowhere, "cellularity emerges" to "encourage coordination." Are you seeing any light that has been shed yet? Later, the personification, assumed goal-seeking, and speculation gets even worse:
The model presented here demonstrates that the modern genetic code evolved most likely by a combination of previously hypothesized forces, involving neutral and selective evolution. Whereas a natural predisposition toward an error-minimizing code is not a necessary condition for an optimized genetic code, neutral evolution may have been an important force in establishing universality. At the same time, selective pressure can provide a powerful impetus for a genetic code to move toward error-minimization and, somewhat surprisingly, also enforce its immutability so as to maintain compatibility with the genome.
Who does the enforcing? Who does the establishing? Who does the maintaining? Who follows an impetus to move toward error minimization? What is an error, anyway, to a mindless molecule? This is crazy, but not crazy enough for the Royal Society to publish it.
They get away with this because it fits the requirement of naturalism: "No intelligence allowed." Within that constraint, they follow Finagle's 6th Rule: "Do not believe in miracles. Rely on them."
Good-bye, RNA World
The authors feel somewhat justified in "envisioning" their make-believe "scenario" on the grounds that "Evidence for such a world [RNA world]... is growing." Too bad this paper came out about the same time that Steven Benner, a veteran origin-of-life researcher, poured cold water on the idea at the Goldschmidt Conference in Florence in August. Here's what he said happens to ribose (an essential sugar for RNA) and other biomolecules when exposed to the watery conditions assumed on the early earth, according to an NBC News article:
The early environment on Earth, however, was challenging to the rise of life as we know it, at least in Benner's view. One of the biggest challenges has to do with the process by which organic molecules gave rise to life's chemical building blocks: RNA, DNA and proteins.
If left to themselves, adding energy to organic molecules just tends to turn them into tar or an oily substance. That's what Benner calls the "tar paradox": How could organic materials ever give rise to biopolymers like DNA?
Science Magazine describes the depressing picture:
However and wherever life began, one thing is sure: Its first organic building blocks, called hydrocarbons, had a number of hurdles to clear before evolving into living cells. Fed with heat or light and left to themselves, hydrocarbons tend to turn into useless tarlike substances. And even when complex molecules like RNA (most biologists' best guess for the first genetic molecule) arise, water quickly breaks them down again.
The RNA-world scenario is so hopeless, in fact, that Benner took the extreme step of claiming that life must have formed on Mars (on dry ground under special conditions), and then got transported to earth via meteors. While some reporters leaped onto the sci-fi suggestion that "We may all be Martians!" (e.g., Space.com), thinking people will surely catch the cry of desperation in such a proposal.
Conclusions
So, even if one were willing to grant the time of day to Jee et al.'s "game theory" notion, Darwinians can't even get the starting materials to play with. It would be more realistic for them to start with balls of tar, and racemic biological gunk broken down by water.
Any way you slice it, the "game theory" approach of these imagineers is an exercise in futility. And that's before even thinking rationally about the problem of the origin of genetic information, discussed in depth in Stephen Meyer's Signature in the Cell.
What a crazy world Darwinism and methodological naturalism (MN) has bequeathed us. The way out is to relax the arbitrary MN rule, to think outside the naturalistic box, and once again, to follow the evidence where it leads. Optimized codes do not "arise" from "frozen accidents." From our universal experience, they are products of intelligent design. That's no game. That's no "scenario." It's reality.
No comments:
Post a Comment