MSNBC has an article titled “Why Scientists Dismiss ‘Intelligent Design'” (go here). In it, Ken Miller argues against ID, and specifically against my claim that undirected natural causes cannot generate specified complexity in biological systems. His argument focuses on the evolution of nylonase:
The nylon problem
There is a way to settle this, however, because like Behe’s irreducible complexity, the concept of specified complexity can also be tested.
“If Dembski were right, then a new gene with new information conferring a brand new function on an organism could never come into existence without a designer because a new function requires complex specified information,” Miller said.
In 1975, Japanese scientists reported the discovery of bacteria that could break down nylon, the material used to make pantyhose and parachutes. Bacteria are known to ingest all sorts of things, everything from crude oil to sulfur, so the discovery of one that could eat nylon would not have been very remarkable if not for one small detail: nylon is synthetic; it didn’t exist anywhere in nature until 1935, when it was invented by an organic chemist at the chemical company Dupont.
The discovery of nylon-eating bacteria poses a problem for ID proponents. Where did the CSI for nylonaseÃ¢â‚¬â€the actual protein that the bacteria use to break down the nylonÃ¢â‚¬â€come from?
There are three possibilities:
**The nylonase gene was present in the bacterial genome all along.
**The CSI for nylonase was inserted into the bacteria by a Supreme Being.
**The ability to digest nylon arose spontaneously as a result of mutation. Because it allowed the bacteria to take advantage of a new resource, the ability stuck and was eventually passed on to future generations.
Apart from simply being the most reasonable explanation, there are two other reasons that most scientists prefer the last option, which is an example of Darwinian natural selection.
First, hauling around a nylonase gene before the invention of nylon is at best useless to the bacteria; at worst, it could be harmful or lethal. Secondly, the nylonase enzyme is less efficient than the precursor protein it’s believed to have developed from. Thus, if nylonase really was designed by a Supreme Being, it wasn’t done very intelligently.
The problem with this argument is that Miller fails to show that the construction/evolution of nylonase from its precursor actually requires CSI at all. As I develop the concept, CSI requires a certain threshold of complexity to be achieved (500 bits, as I argue in my book No Free Lunch). It’s not at all clear that this threshold is achieved here (certainly Miller doesn’t compute the relevant numbers). Nor is it clear that in the evolution of nylonase that anything like pure neo-Darwinism was operating. Instead, we see something much more like what James Shapiro describes as “natural genetic engineering” (go here). And how do systems that do their own genetic engineering arise? According to Shapiro, Darwinism (whether neo or otherwise) offers no insight here.
Let’s look at nylonase a bit more closely. Nylonase appears to have arisen from a frame-shift in another protein. Even so, it seems to be special in certain ways. For example, the DNA sequence that got frame-shifted is a very repetitive sequence. Yet the number of bases repeated is not a multiple of 3 (in this case, 10 bases are probably the repeating unit).
What this means is that the original protein consisted of repeats of these 10 bases, and since it is not a multiple of 3, it means that these 10 bases were translated in all three possible reading frames (the second repeat was one base offset for translation relative to the first repeat, and the next was offset one more base, etc). Moreover, none of those reading frames gave rise to stop codons. Since the 10-base repeat was translatable in any reading frame without causing any stop codons, the sequence was able to undergo an insertion which could alter the reading frame without prematurely terminating the protein.
Actually, the mutation did cause a stop codon; but the stop codon was due not to frame shift but to the sequence introduced by the inserted nucleotide. Simultaneously, the mutation introduced a start codon in a different reading frame, which now encoded an entirely new sequence of amino acids. This is the key aspect of the sequence. It had this special property that it could tolerate any frame shift due to the repetitive nature of the original DNA sequence. Normally in biology, a frame shift causes a stop codon and either truncation of the protein (due to the premature stop codon) or destruction of the abberant mRNA by the nonsense-mediated decay pathway. Nonetheless, the nylonase enzyme, once it arose, had no stop codons so it was able to make a novel, functional protein.
Most proteins cannot do this. For instance, most genes in the nematode have stop codons if they are frame-shifted. This special repetitive nature of protein-coding DNA sequences seems really rare; one biologist with whom I’ve discussed the matter has never seen another example like it. Maybe it’s more common in bacteria. Thus, contrary to Miller, the nylonase enzyme seems “pre-designed” in the sense that the original DNA sequence was preadapted for frame-shift mutations to occur without destroying the protein-coding potential of the original gene. Indeed, this protein sequence seems designed to be specifically adaptable to novel functions.
There is something very special about the nylonase host gene that isn’t true of most genes in general and gives it much greater evolvability. As an aside, the function of the original gene (before it mutated into a nylonase) appears unknown (I’d be grateful for any insight here). The original paper suggested that the host gene was unlikely to encode a functional enzyme on account of lacking the amino acids normally found in active enzymes, so maybe it played some structural role that was not critical for the cell (no mention was made whether the host gene was a duplicate).
Here is a reference to the original paper: “Birth of a unique enzyme from an alternative reading frame of the pre-existed, internally repetitious coding sequence”, Susumu Ohno, Proc. Natl. Acad. Sci. USA, Vol. 81, pp. 2421-2425, April 1984.