Evolution's Credibility Problem (part 2)
(Part 2 in a 3 part series)
At this point I could mention the vast range of astonishingly complex things that evolution is said to have delivered (e.g., sentient life, metamorphosis, sexual reproduction, symbiosis, convergence, molecular machines), but none of these things is really much more incredible than the sub-cellular biochemical changes on which they each depend. Since ignorance is the chief ally of credulity, let me begin with a short lesson in cellular biology.
Nearly every structure in the cell and every chemical reaction that it spawns is dependent upon one or more large molecules called "proteins" (proteins facilitating chemical reactions are called "enzymes"). These molecules are composed of chains of smaller amino acid molecules (out of a possible choice of 20 amino acids), which are folded into various functional arrangements. Only a miniscule number of all possible arrangements of amino acid chains could serve a functional purpose in organisms. The arrangement of the final protein is dependent upon the position of the individual amino acids in the chain, since the right ones need to be in place to provide the necessary cross-linkages in the folds, as well as to result in the right group of chemicals being positioned together to form reaction sites. The final form is much like a custom crafted tool that is designed for a certain task.
SIDEBAR - Most proteins contains at least 100 amino acids. Given that there are 20 possible amino acids for each position within the protein chain, the total possible combinations is 20^100. That works out to a number close to 1 with 130 zeros after it. By comparison, the number of atoms in the entire universe is only around 10^80 (1 with 80 zeros). The largest proteins contain more than 2000 amino acids.Amino acids are assembled into these important sequences by organelles called ribosomes. How do the ribosomes know what arrangements of amino acids to link together in order to make a desirable protein? They work off of a template provided to them by messenger RNA molecules (mRNA). So, how does mRNA arrive at its templates? It gets them by transcribing them from the DNA library. The buck stops at the DNA.
DNA is an unfathomably long double-helix shaped molecular ribbon whose primary job is to store blueprints for proteins. The design instruction for every protein the organism makes is found in some subsection of the DNA. These instructional units are called genes, and they have been likened to a language, with each "letter" representing an amino acid. Each gene would then be informationally equivalent to anything from a long sentence to a page of text.So, how does this relate to evolution? For an organism to change form or acquire some new function, from the largest to smallest scale, it will require one or more new or altered proteins to do the trick. And since protein design is dependent upon the DNA, then DNA gene sequences must be either added or modified to pull this off. Since there is no established mechanism to introduce such changes evolution is dependent upon chance corruptions or copy mistakes (mutations) to deliver these "enhancements."
Mutations are not very common occurrences, since the cell contains a host of attendant proteins whose jobs are to proofread copy results, repair incorrect gene sequences, and reconnect broken strands of DNA. However, the occasional error does happen. This is the heart and soul of the biological change upon which natural selection is said to do its work: weeding the bad changes from the good, and favoring the better designs over the merely adequate existing designs. The problem now for evolution is to come up with beneficial mutations, but even evolutionists will admit that it is far more probable that a mutation be deleterious than desirable. Establishing exactly how improbable is the playground of heroic denial.
To be honest, it is not a precise statistical problem because there are some flexibilities and things as yet unknown in the biochemistry of the cell. However, there is enough that is understood to paint a very black picture for evolutionary probabilities. Let us take as an example a fairly "simple" problem that one bacterium supposedly faced in its evolutionary march toward complexity.
E. coli's primary food source, as with most organisms, is glucose. At some point in its history it would have encountered lactose (milk sugar) and would be at an advantage if it were able to process this, especially when glucose was not available. For this trick it would need an enzyme such as beta-galactosidase (which it does indeed employ) in order to convert lactose into glucose for its direct use. For now, let us assume that evolution needs just this one thing to make a big difference in this one organism. What, then, would it take to get such a thing?
Remember, in order to construct a protein we need a gene to describe it. This means that some mutation(s) must occur to arrange a stretch of DNA nucleotides into the necessary blueprint for beta-galactosidase, or something like it. Just to be generous, let's say that it is possible to make such a protein out of just 100 amino acids. So, how many possible arrangements of the amino acids could make something to do the trick? Just one? One hundred? One million? Let's be generous and say that one trillion trillion different arrangements (one with 24 zeros after it) could make a useful protein.The problem is that compared with the number of possible combinations for the 20 amino acids in 100 various positions, this still means that there is not enough time in all of cosmic history to arrive at one of our arrangements, even if we convert every atom in the universe into copies of genes and randomly rearrange their sequences once per second! And the issues only begin here.
posted by Paul @ 1:00 PM
9 comments
9 Comments:
The issue that really starts here though is the question of what is a realistic way of framing the scenarios that are discussed. Assuming that evolution works by throwing together proteins from randomly selected amino acids and then pointing to how unlikely this is, fools nobody. It is a straw man.
Psio,
I am not "assuming that evolution works" like this, I'm merely pressing the claim made by evolutionary theory itself. Here are a few random examples:
From the Natural History Museum of London we read this:
Natural selection is a critical aspect of the evolutionary process, but it is not the whole story. Evolution depends on there being a diversity of living things for natural selection to act on. The force that creates this diversity is called mutation.
Mutations are random alterations in our genes, the result of genes failing to copy themselves properly or exposure to radiation or other chemicals (called mutagens). (Source)
From Berkeley University:
[A mutation is a] change in a DNA sequence, usually occurring because of errors in replication or repair. Mutation is the ultimate source of genetic variation. (Source)
And from the chief go to website for supporters of evolution, Talk.Origins:
Adaptation is brought about by cumulative natural selection, the repeated sifting of mutations by natural selection.
. . .
The cellular machinery that copies DNA sometimes makes mistakes. These mistakes alter the sequence of a gene. This is called a mutation. (Source)
I have no desire to be disingenuous about this and to burn strawmen. I am sure the real man is flammable enough. The problem with evolutionary advocacy is that the emphasis is often placed on "natural selection," which is less objectionable, while the agent of change, mutation, is pushed into the background or simply discussed in very broad and optimistic terms. I am merely making an attempt to take a very close look at what is being proposed.
Paul,
With respect I think you have sidestepped my point there. None of the examples you cite are problematic as far as evolution is concerned. Mutation is simply not a big deal, it has not been pushed to the background.
I am at a loss, then. It seemed to me that you were questioning the idea that evolution was driven by the random scrambling of amino acids to assemble proteins. Since my point was that the arrangement of proteins is dependent upon DNA nucleotide sequences (genes), and these are arranged by means of mutations, then it seemed as though you were disputing that evolution was driven by random mutations. To that end I provided some quotes in support that fact.
Since things like "random alterations" and "copy mistakes" are principally blind to any design objectives, and since there are profound functional lacunae between arrangements of chemically active (much less beneficial) proteins, then I cannot imagine how one would see such mutations as "not a big deal," unless it could be demonstrated that favorable mutations are in reality a dime-a-dozen. That has most certainly not been done. In fact, efforts to produce favorable mutations have been extremely thin. "Thin" both in the number of researchers making the attempt (how odd that there are not more), and "thin" as in yielding unimpressive results. Please note that this is not as problematic a task as it might appear, since it is possible with bacteria to simulate the equivalent of 1 million years of mammalian evolution in just a matter of decades.
Paul,
It seems to me that you have made an implicit assumption that 1 mutation = 1 amino acid in the sequence. Clearly it cannot work this way. Mutations are random but they occur within a context. It is this that you are not taking into account.
Psio,
In my understanding of this I have not made that assumption, but perhaps I have mistakenly given the impression that this is so. I certainly did suggest that there is a one-to-one relationship between codons in the DNA, which may be mutated, and their associated amino acids. But since I did not go in to how mutations typically proceed I don't think you can claim I've made an explicit error.
I think you are mistaken about mutations occurring in a "context." The only context is the DNA (though non-inheritable errors can also occur in transcription and translation). Mutations either occur as corruption of the DNA via chemical or radiation assaults, or they occur as mistakes in the replication process. Both of these things occur in a way that is blind to the contents and partitioning of the DNA, unlike when DNA is transcribed into mRNA (which has its eye out for start and stop codons as well as promoter sites). That is to say, mutations do not target their damage to open reading frames, introns, exons, or any other domain of the DNA. The mutations can occur anywhere at all and may be small or large in scope.
Most commonly, the mutations do happen to involve single nucleotides [3 nucleotides = 1 codon, which may define an amino acid]. Either an existing nucleotide is changed, a nucleotide is removed, or one is inserted. Changing a nucleotide will either have no effect at all (since most amino acids are defined by more than one codon arrangement); or it will code for a different amino acid; or it may spoil start, stop, or promoter codons (thus ruining the gene); or it may make a new start or stop codons, also ruining the original gene.
If a mutation causes one or more nucleotides to be inserted or deleted, unless done in multiples of three, it will cause a "frameshift" on the gene. This will completely change all of the codons from that point downstream in the gene. Besides yielding amino acid potluck, it is guaranteed to affect the stop codon and will quite often introduce another new stop codon somewhere in the frameshifted mix. While this may be the most effective type of mutation for generating a long stretch of "new" sequences, it is also the most devastating to the DNA sequence that it supersedes.
So at worst, trying to generate a new gene is like trying to change one paragraph into another one, one or two letters at a time (within the whole book), without allowing the text to pass through a verbally meaningless stage. And at best, it is like pouring a whole box of letters out on the floor and hoping they will land in a meaningful arrangement, punctuation and all.
Of course, there are some other less common types of mutations, but they either cause worse damage to the original material or they simply copy sections of what is already there. They do not, however, offer much help in adding new information.
So it seems to me that, far from being like a random insertion into a fixed text we have many possibilities including frame shifts, duplications, deletions and insertions into the mechanism that generates the text. Most of these will be deleterious but since we have a ratcheting system of accumulating beneficial change and many millions of mutations happening within each generation of a population, it is not clear at all that you have made the case that there is not enough here to drive evolution.
Psio,
I think you are simply assuming that lots of mutations and time automatically means that anything's possible. What if I gave you 20 dice and had you roll them once per second for 1 million years (and I made you immortal)? Do you think you could roll all sixes in that amount of time? Sounds doable, but you'd only get in 3.15 x 10^13 rolls, and there are 3.6 x 10^15 possible roll combinations! You'd need more than 100 friends to help in order to have a chance at success.
The problem with tweaking nucleotides here and there, or randomly scrambling segments, is that you're playing similarly random dice games with even more astronomical odds. And you're rolling the dice on top of your densely packed game table, which endangers the whole the game. It's like trying to roll 20 (I'd say more) sixes without disturbing your existing sets of 20 fives, fours, threes, etc.; or if you do disturb them, it must yield an equally interesting arrangement. Getting even one good roll is a miracle, much less getting the kinds you really want, much less getting them in spades and "accumulating" them.
This reminds me of back of the envelope calculations people use to verify the existence of paranormal phenomena. Superficially plausible but light on context and actual data. Of course you can construe the way it works in an infinitely many wrong ways that will yield these unfeasible numbers.
Post a Comment
<< Home