Junk DNA and creationist lies

Creationists deny that ANY of our DNA could be non-functional, leftover junk that has no selective effect. But the overwhelming evidence has long been against this notion that no real biologist takes seriously, and a new study drives this home even further.

One of the common tropes you hear among modern creationists is the denial of the idea that there is any non-coding DNA, or “junk DNA.” To them, the idea that a large part of the genome is simply unread leftovers, carried along passively from generation to generation without doing anything, is clearly a contradiction with the idea of an “Intelligent Designer.” So the Discovery Institute and numerous other creationist organizations that are actually sophisticated enough to recognize the issue (including Georgia Purdom of Ken Ham’s “Answers in Genesis” organization) keep spreading propaganda that “junk DNA is a myth” or “every bit of DNA has function, even if we don’t know what it is.” Moonie Jonathan Wells, who has written crummy books misinterpreting fossils and embryology, wrote a whole book denying the subject—even though he hasn’t done any research in molecular biology since 1994. Do they actually do any research to explore this topic, or trying to test the hypothesis that all DNA is functional? No, their labs and their “research” are not that sophisticated. Instead, their entire output on the topic (just as in every other topic) is based on cherry-picking statements of the work of legitimate scientists, quote-mined to distort the meaning of the original scientific publication. Either they don’t understand what they are reading and their confirmation bias filters screen out all but a few words that seem to agree with them, or they are consciously lying and distorting the evidence—or both.

First, some background on the issue. Before 1966, nearly all biologists were “panselectionists,” convinced that every part of an organism was under the constant scrutiny of natural selection, even if we couldn’t detect it. Even in the 1970s, I had professors in evolutionary biology at Columbia University who were hard-core panselectionists, and could not imagine the possibility that nature was not very efficient, but could carry along structures from one generation to the next that either had no function, or were suboptimal in their function (as Stephen Jay Gould had been advocating). But as early as 1966, Lewontin and Hubby (using the then-new technique of gel electrophoresis, which has long since been replaced by direct DNA sequencing) showed that the variability in the protein sequences in many organisms was far in excess of what was needed to explain their anatomical complexity, and that there was no correlation between complexity and genetic information: there were simple worms with huge genomes, and complex organisms with small genomes.

I vividly remember the next step in the debate, because it was raging when I was in grad school in the 1970s. As soon as the basic nucleotide sequence that specified the type of protein was decoded, it became apparent that a large number of mutations could be “silent.” Typically, in the three-nucleotide “codon” that determines a given protein, the first two “letters” (nucleotides) determined which protein would  be produced, but the nucleotide in the third position could be any of the four possibilities (adenine, guanine, cytosine, and tyrosine or uracil) and it would make no difference—the same protein would result. Thus, a mutation in the third position is usually invisible to natural selection, and could randomly change from one condition to another without any phenotypic effect. Also, at this time scientists were first discovering evidence of “molecular clocks”, which would only work if a large portion of the genome were not under the supervision of natural selection, but operating like a ticking “clock,” randomly changing without any external modification. By the late 1970s, “neutralism” (the idea that a lot of the genome was selectively neutral) was all the rage in evolutionary biology, spearheaded by Motoo Kimura’s book on the topic.

The genetic code. The first two "letters" in the triplet sequence normally is sufficient to specify a particular protein; changes in the third "letter" make no difference in protein, and thus are selectively neutral.

The genetic code. The first two “letters” in the triplet sequence normally is sufficient to specify a particular protein; changes in the third “letter” make no difference in protein, and thus are selectively neutral.

Fast-forward to the present day, and as more and more molecular studies have been undertaken, more and more evidence of non-coding DNA has been discovered. These discoveries come from three major lines of evidence. First, as Lewontin and Hubby noticed almost 50 years ago, there is no correlation between the amount of genetic material and the complexity of the organism. As one textbook put it:

The DNA content per cell also varies considerably among closely related species. All insects or all amphibians would appear to be similarly complex, but the amount of haploid DNA in species within each of these phyla varies by a factor of 100. The same variation in DNA content per cell is common within groups of plants that have similar structures and life cycles. For example, the broad bean contains about three to four times as much DNA per cell as the kidney bean.

Other examples abound: one species of deer has 20% more DNA than its closest relative. Even one species of fugu (pufferfish) has almost 100x as my DNA as another. Clearly, since there are no significant differences between such pairs of species that could account for so much variability in DNA content, much of it must be junk. This is one line of evidence that the creationists never address, since they have no explanation for it—and nor does any legitimate biologist.

Second, it’s possible to delete some of this repetitive non-coding junk sequence and nothing happens. In 2004, Nabrega et al. deleted almost 3% of the mouse genome that appeared to be repetitive and non-coding, and the mice continued to reproduce with no ill effects. If this DNA were functional, how could the mice keep on reproducing without it?

But the most convincing evidence has come from our greatly improved understanding of entire DNA sequences, and what they consist of. Creationists play this game of “we just don’t know what most the DNA codes for, so we can’t rule out that it has a function,” but this has been a lie for decades. There are many different kind of DNA sequences that are clearly non-functional. These include:  1) pseudogenes, which look like they were once functional DNA, but have lost the ability to be expressed; 2) Transposons, or “jumping genes,” which can jump from one part of the DNA to another and yet are not expressed; 3) SINEs (short interspersed nucleic elements) and LINEs (long interspersed nucleic elements) which are segments of DNA stuck in the middle of a coding sequence that have no function or ability to code for proteins; 4) highly conserved non-coding non-essential DNA, which is very consistent in the sequences of many organisms suggesting that it is important, yet can be removed with no effect whatsoever (Westphal, 2004); and 5) repetitive DNA, which repeats the same sequence over and over again hundreds of times, and none of this repetitive genome seems to code for anything. Perhaps the most interesting and surprising of all of these junk sequences are endogenous retroviruses (ERVs). These are gene sequences of retroviruses that once infected us by inserting their DNA into our genome, but are no longer active. Instead, every time one of our cells divide we make new copies of this “fossil DNA” from a long-ago viral infection and carry it on through millions of generations. Clearly, these “DNA fossils” hiding in our genome no longer code for the virus, or for anything else—they are clearly “junk” that we passively carry around with no ill effects. In the few cases where the retroviral DNA can be re-activated, it  usually causes disease or other bad effects. Either way, it is of no comfort to creationists.

Naturally, creationists have trouble addressing all this evidence, and instead cling to any bit of biology that seems to support their belief system. In 2012, the ENCODE Consortium studies were published, which made a big splash because they posited that maybe 80% of the genome did produce some kind of protein. Naturally, both the media and creationists jumped on this to confirm their belief that all of the DNA was functional. But although this still concedes that at least 20% of the DNA is clearly non-coding and no comfort to creationists and the “intelligent design” idea, the creationists haven’t noticed this, but proclaimed that they had been vindicated. Sorry, but it turned out that the ENCODE studies were too good to be true. Graur et al. (2013) have just published a paper that completely demolishes their work, and reaffirms that indeed most of the genome (at least 90% of it, perhaps as much as 98%) is non-coding. P.Z. Myers goes over the study in detail. The salient point is that all the ENCODE study managed to show is that some of the genome called “junk” codes for a protein. What they didn’t show is whether these random isolated proteins actually are part of a functional biochemical pathway, or lead to phenotypic consequences. In fact, if a protein results from “junk DNA” but doesn’t do anything, it’s still “junk.”

But if I’ve learned anything from the battle with creationists, not only will they continue to misinterpret the ENCODE Consortium studies to argue that all DNA is functional, but they will ignore the debunking by Graur et al. (2013) and all future studies with the same effect. Creationists have consistently demonstrated that their confirmation bias filters are very strong, and their habit of cherry-picking and distorting the meaning of real scientific studies is part of their DNA.



  • Graur,D., Zheng, Y., Price, N., Azevedo, R.B.R., Zufall, R.A., and Elhaik, E. 2013. On the immortality of television sets: “function” in the human genome according to the evolution-free gospel of ENCODE. Genome Biology and Evolution (published online)
  • Nabrega, M. Y.  Zhu, I. Plajzer-Frick, V.  Afzal and E. M. Rubin. 2004. Megabase deletions of gene deserts result in viable mice. Nature 431:988-993.
  • Westphal, S.P., 2004. Life still goes on without ‘vital’ DNA, NewScientist No. 2450, P.18.