Monday, 2 July 2012

Addition of Genetic Information Redux: A Critical Response to a Critical Response

A few years ago ( when my blog was a bit more active than it is today) I wrote up a post rebutting the old creationist canard that evolution requires "new" genetic "information" to be added to a genome, and, furthermore, that such a process is impossible and consequently, evolution is false. Recently, a LiveJournal user by the name of Jair_greycoat wrote up a response to my article. After reading it, I think that there are a number of points that I should clarify and some misconceptions that should be corrected. Such a task is too much for the LiveJournal comment section, so I've written up my reply to the criticisms below.

"Why would creationists claim that beneficial mutations are rare or impossible? Didn't the author of the above quote just provide a real, observable example? The only claim I have heard from creationists regarding this is that all, or nearly all, such beneficial mutations are a result of a loss in information, not a gain. I have not personally heard them claim that the loss of information cannot be helpful. We do have a saying, "ignorance is bliss." If a person's body is "ignorant" of alcohol, then they have "bliss"--they can't get drunk, or so I've heard. Nevertheless, the author of the quote agrees with the creationists that the mutation is a loss of information."

Why would Creationists claim such a thing? Your guess is as good as mine. The scientific literature is replete with examples of mutations which confer a beneficial or improved phenotype. The classic example of sickle-cell anaemia is one, as is lactase persistence in European populations (see here and here). Creationists, nonetheless, are not always ones to let reality get in the way of their religious beliefs, and routinely claim that beneficial mutations do not – and cannot – occur. The argument is quite common in the Creationist literature; see this article from for example, "The Myth of Beneficial Mutations", which outright claims "The bottom line is that mutations always weaken an organism". Other examples of this can be seen here and here and here. The claim comes up so often that even Answers in Genesis had to tell people to stop using it.

The claim that beneficial mutations only come about by a loss of information, as you mention, is another claim that comes up - one that is equally false. The example of sickle-cell anaemia as mentioned above is not due to a "loss of information" (which I take as to mean a deletion in a gene) but rather due to a change in a single nucleotide in the gene – from an A to a T. This change results in a protein that is altered in one amino acid, and confers resistance to malaria. Another example is the evolution of lactulose metabolism in E.coli, which occurred not through the "loss of information" but rather from a genetic recombination within a previously existing gene; that is to say, the "information" (for the lack of a better term) was rearranged and gave rise to a novel, beneficial function. Or how about this interesting example: some placental function in vertebrates is due to a proviral gene that is integrated into our genome – again, a gain of genetic "information", rather than a loss of it. The point I'm trying to make is that for every example you can point to where a beneficial phenotype has arisen due to a genetic deletion, you can also point to one that was caused by a gain of "information".
It is one facet of this "gain of information", namely diversification of gene function via genetic duplications, that the rest of my article aimed to address. The creationist claim I was rebutting was that a "gain of genetic information" is impossible, and I went on to provide a mechanism whereby it can occur.

"The author then says that the term "information" is too vague and not defined by creationists. Personally I've never had a problem with the word, I think it means exactly what it sounds like it means."

Perhaps he does not realize that the word "information" has multiple specific meanings in science. When a creationist mentions "information", are they meaning actually physical information? In which case, are they referring to classical information or quantum information? Or maybe they mean information in the technological sense, as in "instructions" or "code"? A loss or gain of "information" takes on very different meanings depending on which sense of the word you're using. "Information" is a rather clumsy word to use when describing genetics for this reason, and it is not used widely in the biological community. Its use in reference to genetics seems to be restricted to Creationists.

"Why would creationists claim that genes are never added to a genome? Of course genes are added; I think we can agree on that. What I have heard creationists deny is that these new genes contain new and original content, or that they are capable of transforming into new and original content that is not simply random garble."

See above. Creationists claim that because they are either ignorant of the facts or refuse to accept them.

"Now I realize that the writer is probably using the words "gene" and "information" interchangeably."

Again, refer to my explanation of the problems with using the word "information" above. I used "information" throughout my original article because it is the term that Creationists throw around all the time. I use it in quotes to show that it is an ill-fitting word be using in such a context. I used "information" to refer to both genes and gene sequences, since that is what I think Creationists mean when they throw the term around.

"Then the author goes on to quote Ross Hardison. As I see it, the quote is essentially a hypothesis about the evolution of a haemoglobin gene. I don't have any questions about it except this part: "In this way, the two genes that started out identical acquired sequence differences and later, functional differences." But how do these mutations become functional differences rather than just causing everything to fall apart, as I would expect from entropy?"

To answer this, you must really think about just what a mutation means to the gene it's situated in. A mutation, whether it is a deletion, addition or just a change in the sequence, carries with it the risk of altering the sequence of the protein for which it codes. Proteins get their function from their shape, and their shape is directly (more or less) determined by their amino acid sequence. As mutations accumulate in a duplicated gene, the chance of these mutations altering the function of the protein it encodes rises. Of course, it is not guaranteed that these alterations will be of any benefit, but that's where natural selection plays a role. In those instances where the mutations are harmful, the gene is less likely to be passed on. Those that confer some benefit are more likely. In this manner, over generations, the net effect becomes a positive one. When we look at the haemoglobin gene today, we're looking at it with a bias – we only see the mutations which were passed on through the generations and not the harmful ones that were weeded out. 

Have you ever played the game Yahtzee? In that game, you have to roll five dice, and you're scored based on the combination of numbers that result. The best score you can get on a turn – the eponymous Yahtzee – is to get a 6 on all five dice. The chances of getting a Yahtzee are pretty low, only 1 in 7776. But the game lets you roll the dice three times, and you're allowed to select the dice you want to keep between rolls. So if you roll 6 on two dice on your first attempt, you can keep those and only roll the remaining three on your second try. In this manner you greatly increase your chance of scoring a Yahtzee; it now becomes 1 in 22 (see here if you're interested in the math behind it). Natural selection works in exactly the same principle. The odds of getting a beneficial new function out of random mutations is low without selection. But selection allows those good mutations to be saved and the bad ones to be discarded. Once you apply selection, novel functions can arise quickly and easily. 

I should also point out here that even a loss of function or a reduced function is still a functional difference. Any mutation that is not neutral or silent, by definition, results in some functional difference. The question, then, should not be "how do mutations become functional differences" but rather, "how do these mutations result in something that is beneficial rather than deleterious", the answer to which is natural selection.

One other point that I would like to make before continuing is regarding the author's use of the word "entropy". I have seen this word tossed about by Creationists more times than I can count, and in almost all the cases, they are not using the word in the correct sense. Creationists are often quick to cite the Second Law of Thermodynamics as "Over time, the amount of entropy in a system increases", and then claim that evolution violates this principle as it requires that over time, entropy must decrease, viz. genomes becoming more ordered rather than degrading into non-coding gibberish. The Second Law of Thermodynamics is no obstacle to evolution, however, as the definition of the law cited by Creationists leaves out a pivotal point: that it applies only to a closed system, that is, a system where there is no flow of energy into or out of it. Biological systems are, of course quite open, with energy flowing freely into and out of them. Genomes are under no obligation to fall victim to increasing entropy. 

"Again from the main article, I quote: "I can already hear the cries of the creationists. "But," they proclaim, "this doesn't show evolution at all, for the different haemoglobin genes are still all the same kind!" (Oh how I hate that dreaded "kind" word)." I'm going to step out and say that I have never heard any serious creation-scientist speak or write a sentence like that one. It does not even make sense in context. Why would a creationist talk about "kinds" with regard to genes in the first place? "Kinds" as far as I have heard it used, is a word used by creationists to refer to different species. It would make more sense, I think, if the sentence was written as: "But," they proclaim, "this doesn't show evolution at all, because it is only a story which doesn't explain the fundamental difficulty! How can random mutations produce meaningful structures which help a species' survival?""

I cannot comment about how familiar the author is with the claims of "serious creation-scientists" but the "kind" argument is one that I've heard a mindboggling number of times. It seems to be a favourite of Kent Hovind and Ken Ham. They argue that "no dog ever gives birth to a cat" (which, if it occurred, would be a blow against evolution, not in support of it) and that's because dogs are one "kind" and cats are another "kind". The major problem with the "kind" talk is that Creationists never use the word in a consistent manner. The author claims that they use "kind" synonymously with "species", which is blatantly false. Are "dogs" and "cats" species? Creationists will claim that Drosophila melanogaster is a "kind", but also that flies in general are a "kind". They use the word to refer to whichever biological taxon is convenient to them at the time, whether that be a species, genus, family or other taxon. It is in this sense that the hypothetical Creationist response that I proposed makes sense. The haemoglobin example I used show how new, but similar, haemoglobin genes evolved. The likes of Ham and Hovind would classify these as the same "kind" in the same way they classify dog breeds as "kind" (note that I'm not saying that such a comparison is justified, only that it is one that I imagine Creationists would make). The remainder of the paragraph the author quoted then went on to explain how new different functions can arise.

"The article explains that new structures--new content in the genetic code--can come from numerous mutations over time. Bad mutations would be weeded out by natural selection, and good ones kept. This made sense to me--on the surface--collect enough mutations over enough time, and there is the possibility of hitting a combination containing survival value simply by chance. I'll try to explain why I think it doesn't make sense on a practical level later in this entry."

Here the author demonstrates that he does understand the process of natural selection, if only superficially. This makes me wonder why he seems to have difficulty understanding how this principle can be applied to duplicated genes evolving novel functions.

"I don't know exactly what percentage of mutations creationists claim are harmful. However, isn't any mutation that isn't beneficial, a potentially harmful one? Especially when you have a large number of mutations, each being by itself neutral or nearly so; but when many such mutations pile on, it doesn't matter--the overall original function of the gene is destroyed, because it no longer contains the original instructions for whatever structures it previously needed to survive...Why would only one of the copies be mutated? Wouldn't both of them mutate at the same rate?"

As mentioned above, Creationists alternately claim that all mutations are harmful, or that most mutations are harmful. The reality is that the vast majority of mutations are neither harmful nor beneficial. Most mutations are neutral, that is to say, they have no affect on the gene at all. Since the genetic code is redundant (i.e. there are many nucleotide codons which code for any given amino acid), most mutations won't affect a protein's sequence at all. Furthermore, similar codons code for amino acids that are chemically similar, so a mutation that does alter the protein sequence will not necessarily alter the protein's function. These neutral mutations confer no advantage or disadvantage on an organism and are therefore invisible to the eyes of natural selection. If a gene were to acquire a large number of neutral mutations, it's affect on the gene would be minimal precisely because they are neutral! The original function of the gene would not be destroyed at all. 

As for mutation rates: yes, it would be safe to assume that both the genes would have the same mutation rate. However, it has little bearing on the ultimate fate of the genes. The reason we see many more mutations in one copy of the gene and not the other is because a deleterious mutation in both copies could prove lethal. Any organism that had both copies mutated in such a way would die and those versions of the genes would not get passed on to successive generations. However, it that same deleterious mutation occurred in one copy and not the other, this would not affect the organism since it still has an original, functional copy remaining. The other copy is free to accumulate mutations as long as there is still an original, functional copy remaining. 

"But in order for a mutation to be considered helpful to a creature's survival and therefore selected by nature, doesn't it have to be part of a fully functional structure from the very start? If not, in what way is the mutation helpful? I believe that even if the mutation is comparable to the first steps in building a genetic program, until that program is finished and complete, it will be a drain on the organism's resources, and therefore harmful. I should think then, that even mutations that could in the future be beneficial (those that are not merely losses of information) would be weeded out by natural selection. Hence why the article I quote from does not make sense to me in this respect. "Fundamentally, this type of evolution requires natural selection to step into the picture and choose between variations--which organism is most fit. But isn't natural selection a blind process? At least according to atheists, I have heard that it is. How can a blind process select for mutations that have the future potential to be part of new functions and structures, yet have no survival value (or worse, are a net drain on the organism) in the present?"

What the author presents here is his own version of the Irreducible Complexity (IC) argument. "What use is half a wing?" the argument often goes. The examples that are usually brought up – the eye, the bacterial flagellum – have been debunked countless times. I could write an entire blog entry on the problems with the Irreducible Complexity argument, so forgive me for being terse here. The major flaw in IC is that it does not account for the gradual adaptation of one biological system for a different, novel function. What good is half a flagellum to a bacterium, you ask? Well, it works great as a Type III Secretion system. Sure, if you removed any of the flagellar components, it might cease to function as a flagellum. But that ignores the fact that the "half a flagellum" may have some other evolutionary adaptation entirely. This can be said of any supposedly "irreducibly complex" system. Interested readers can read more about the problems with the IC argument at TalkOrigins.

"If by "no new genetic info" the author means "no new and original content" rather than copies of previously existing genes, then the reason the claim is so common is because (as far as I know) there is no explanation from evolutionists as to how new and original information can come from mutations and natural selection."

If this is the case than the author has either misunderstood the mechanism that I have described in my article or is unfamiliar with the criticisms of Irreducible Complexity. Again, I refer to the TalkOrigins article I linked to above.

"What about the creation-scientists who are also part of the scientific community and who probably say quite loudly that gene duplication is not a sufficient mechanism for evolution? This quote just feels like the author of the article is intentionally ignoring them. This probably bothers me even more than the article's explanation for gene evolution, because it seems to imply that the author, at some level, does not see creation-scientists as "real scientists". That is just wrong, especially as I have seen plenty of books written by them, and those which I read made plain and simple sense to me. Certainly more sense than the article I quote from made."

The author does get this one thing correct: I don't consider "creation-scientists" to be real scientists. If the author has any particular individuals in mind who he feels are real, practicing scientists, I would love to know who they are. But for the most part, "creation scientists" fit one or more of the following descriptions:
  1. Do not have a degree in a scientific field related to evolution (evolutionary biology, molecular biology, etc.).
  2. Do not have a degree from a recognized, accredited institution.
  3. Do not publish actual research in peer-reviewed scientific journals.
Unfortunately, if they do not match these criteria, then they're not real scientists. It might sound harsh but that's the way science works. Again, if the author has literature from a scientist who holds a relevant degree from an accredited institution that has been published in a peer-reviewed journal, I'm all ears. But until then, I'll give "creation scientists" all the credit they deserve; which is to say: none.