On August 19, Gina Kolata reported in The New York Times that geneticists “have seen a dead gene come back to life and cause a disease.”
According to Kolata, the human genome “is riddled with dead genes, fossils of a sort, dating back hundreds of thousands of years–the genome’s equivalent of an attic full of broken and useless junk,” though some of those genes “can rise from the dead like zombies.”

The Science article proposes a hypothesis in which DUX4 causes FSHD, and people with fewer than 11 D4Z4 repeats make DUX4 that is more stable (and thus more toxic). But the Science article says nothing about “dead genes” coming back to life.
Instead, Kolata quotes Dr. Kenneth Fischbeck of the National Institute of Neurological Disorders and Stroke, who told her that FSHD was due to “an ancient gene stuck on the tip of chromosome 4. It was a dead gene; there was no evidence that it was expressed.” Dr. David Housman, a geneticist at M.I.T., said it might be possible to treat FSHD by “turning off that dead gene.” Dr. Francis Collins, Director of the National Institutes of Health, expressed surprise at the discovery: “To come up with this sort of mechanism for a disease to arise–I don’t think we expected that.” According to Collins, “If we were thinking of a collection of the genome’s greatest hits, this would go on the list.”
But how do Fischbeck, Housman and Kolata know that D4Z4 or DUX4 were once dead and have only recently become functional?
The short answer is, they don’t.
The closest we can come to determining the functionality or non-functionality of DNA in ancient organisms is to compare DNA in living organisms. By the logic of neo-Darwinian theory, random mutations will accumulate in a DNA sequence unless they are weeded out by natural selection, but natural selection can select only for or against a function; it cannot affect a DNA sequence that is non-functional. If two modern species inherited a non-functional DNA sequence from a common ancestor in the distant past, the corresponding sequences will be unaffected by natural selection, and they will have become very different in the descendants due to accumulated mutations. We can turn the logic around: If corresponding sequences in modern species are very similar, then neo-Darwinism implies that natural selection has “conserved” the sequences because they are functional–even if we don’t know what that function is.
In 2007, British geneticists reported that they had compared D4Z4 and DUX4 sequences in humans and other primates with sequences in mice, rats, elephants. They found that the sequences were conserved among these mammals, and they concluded that this “strongly supports a coding function for D4Z4.” Since humans and rats supposedly diverged from a common ancestor over a hundred million years ago, the evidence suggests that D4Z4 has been functional for at least that long.
In other words, the evidence doesn’t support the “dead gene” story, which comes from another source entirely.
When Darwin’s theory of evolution was combined with Mendel’s theory of genetics in the 1930s, the resulting “neo-Darwinism” assumed that genes are the heritable elements that contain the information to make organisms–thus gene mutations could generate the new variations that supply raw materials for evolution. When James Watson and Francis Crick discovered the structure of DNA in 1953, they provided what seemed to be a molecular basis for neo-Darwinian theory: DNA makes RNA makes protein makes us. But when molecular biologists discovered in the 1960s and 1970s that most of our DNA does not code for protein, some people called the non-protein-coding DNA “junk.”
In particular, the non-protein-coding repetitive DNA that makes up almost half of the human genome was relegated to the garbage dump. In his 2006 book The Language of God: A Scientist Presents Evidence for Belief, Collins paradoxically argued that our DNA–especially its repetitive elements, “with roughly 45 percent of the human genome made up of such flotsam and jetsam”–does not provide evidence for belief in God but instead provides “compelling” evidence for Darwinism.
Richard Dawkins, Douglas Futuyma, Michael Shermer, Philip Kitcher, Kenneth Miller, Jerry Coyne and John Avise have also written recent books in which they argue that much of the human genome consists of “junk DNA” that provides evidence for Darwinian evolution–and evidence against intelligent design.
But the notion of “junk DNA” owes more to the historical contortions of neo-Darwinian theory than to biological evidence. In fact, there is now a large and growing body of evidence that Collins, Dawkins, Futuyma, Shermer, Kitcher, Miller, Coyne and Avise are dead wrong on this point–as I will show in my forthcoming book, The Myth of Junk DNA.
And Kolata’s description of D4Z4 and DUX4 as “zombie genes” that once were dead but now are alive is completely unwarranted.