• Darwinian Evolution on a Chip

    One of the things that I’ve really enjoyed about the creationism war is learning new things.  It’s especially fun when you’re talking to a creationist and they trot out an old, worn analogy or claim and you can dump an actual paper on them that shows they are wrong.  It’s almost disturbing how good it feels.

    Something I would like to do here is provide something of a resource to defeating even the uncommon arguments both in creationism, anti-vax, anti-GMO, and other pseduoscience arguments.  This post directly involves Intelligent Design’s poster boy, Michael Behe.

    Behe, at some point, was probably a decent biochemical researcher.  He’d probably never have gotten into Nature or Science, but then not everyone does.  But he could have plodded along, contributing to the knowledge of humanity.  Then he let something bad happen.  He let belief interfere with reality and it’s basically killed his career.  Even his own department has issued a statement saying his work is full of crap (yes, that’s a paraphrase).

    But this post is about a specific claim of Behe’s and a specific refutation of that claim.  I’ll throw in a section on another claim of ID proponents as well, think of it as a freebie.

    Michael Behe, in his bookEdge of Evolution states that it is impossible for four mutations to happen in a gene to result in a improvement in the resulting protein.  He uses the example of resistant malaria, where there are two mutations and states that this is the edge of evolution because there’s <some huge number> probability that these two mutations can’t happen in the same gene at the same time. Let’s see what Behe says:

    The Darwinian magic works well only when intermediate steps are each better (“more fit”) than preceding steps, so that the mutant gene increases in number I the population as natural selection favors the offspring…Yet its usefulness quickly declines when intermediate steps are worse than earlier steps and is pretty much worthless if several required intervening steps aren’t improvements).
    (Michael Behe, The Edge of Evolution: The Search for the Limits of Darwinism, pg. 112, (Free Press, 2007).

    Basically, Behe is saying that if something bad happens, then it’s more likely that the gene is forever broken than it could improve.  Well, I’m here today to show you just how wrong Behe is.  Darwinian Evolution on a Chip is our reference paper today.  This is a very approachable paper, I encourage you to read it.  But here’s the highlights.

    Gerald Joyce and his team set up an automated evolution lab.  They used very small amounts of an RNA enzyme that bonded to a particular substrate. The polymerase enzymes were attracted to RNAs that had bonded with the substrate.  This was the reproduction part. By continually transferring the population of RNA to work cells with less and less substrate, they were able to select for RNAs that were very efficient at bonding.  This is the selection part.

    Now, if you are talking to a well informed (hah!) Intelligent Design proponent, then they will stop you and say “But the experiment was intelligently designed, therefore the results are intelligently designed.”  Which is utter BS if you think about it.  If the results of any intelligently designed system were put into place before hand (front-loading) or somehow known to the designers, then the entire concept of the casino is wrong.  From roulette to pachinko, there are many intelligently designed systems that are effectively unknowable when it comes to results.  Sure, the results are constrained, but so is life.  If a mutation results that allows an organism to utilize nylon waste as an energy source, to be useful that organism must be in a place that has nylon waste.

    In fact, an experiment has to be intelligently designed.  The ability to reduce variables, account for error, and come to a valid conclusion require that an experiment be very intelligently designed.  I usually note at this point that intelligent design proponents don’t do experiments.  So what does that say about them?

    Anyway, back to our RNA on a chip.  The results of this experiment were absolutely amazing.  In 70 hours (500 ‘generations’), the RNA had increased it’s efficiency at bonding to the substrate 92 fold.  Eleven (11) mutations had occurred which were grouped into four broad categories M1, M2, M3, and M4.

    That part right there just blows Behe’s entire argument out of the water. Eleven point mutations which resulted in a 92 fold improvement.

    The RNA started in a solution concentration of 1 micromol and was binding at a rate of .6/minute.  At the end of the experiment (70 hours), the concentration was down 0.005 micromols and the binding rate was above 20/minute.

    Here’s the next awesome part.  Joyce and his team, took samples of each generation and sequenced them looking for the various mutations.  Then they could purposefully insert those mutations to see what affect it had on the binding rate.

    M1, by itself, resulted in a 9 fold increase.  M2, by itself, was about the same.  M3, by itself, resulted in a 24 fold increase.  M4, by itself, resulted in a 2 fold decrease.

    Yes, M4 was a harmful mutation.  By itself, M4 decreased the RNA’s ability to bind to the substrate.  So why was M4 in the final population at all?  Because M4 is what I call a potentiating mutation.  Even though, by itself, it’s bad.  When combined with the other mutations, it rapidly increased the binding rate of the RNA.

    So that neatly defeats part of Behe’s claim as well.  Bad mutations aren’t always bad.  The classic example, of course, is sickle cell anemia.  One sickle cell allele causes a mild form of the anemia, while two alleles results in early death.  So, why does this condition even exist if the homozygous condition kills children?  Because the heterozygous condition provides an advantage if one lives in an area with malaria.  In this case, the heterozygous condition (1 allele) has a higher fitness than either homozygous condition.  Of course, if you don’t live in a high malaria area, then the homozygous non-sickle cell condition is more advantageous.

    So, any time anyone points out these two claims A) that more than two mutations can’t happen to improve fitness or B) bad mutations always result in the breaking of the protein.  Just discuss this article and show them that actual experiment trumps thought experiment.

    Category: CreationismEvolutionScience


    Article by: Smilodon's Retreat