Joined: Oct. 2006
|Quote (phonon @ April 06 2007,10:36)|
As much as he'd like to believe so, this article is NOT pro-ID. It speaks of rational design or intelligent design, but that's not the same as Intelligent Design.
This is biomimetics. Duh. It's called directed enzyme evolution.
From the article:
|Ultimately, the objective is to make proteins perform for us as well as they perform in life.|
And dig this, Dembski:
Oh noes, did we improve upon the work of the Intelligent Designer? How can that be? Well, you always said that Intelligent didn't mean perfect. I guess the Intelligent Designer makes mistakes. Can't be who you had in mind then, can it?
|Rational design has been mainly used in biotechnology to improve the properties (especially thermostability) of natural enzymes.|
Another nice quote for Dembski:
|It is often said that random genetic methods to improve enzyme properties “rely on simple but powerful Darwinian principles of mutation and selection” (Johannes and Zhao 2006). We agree. It is also said that “every protein has become adapted by step-by-step improvement and refinement of its function over millions of years” (McLachlan 1987). The present theories, however, only partly explain the protein diversity, although a recent study (Poelwijk et al. 2007) shows that even the key-lock dilemma can be resolved by the Darwinian approach when the operation field for random search is within the same protein family, and the new key-lock pair closely resembles the original (ancestral).|
Ohno, those theories from Ohno are decades old.
|Gene duplication and subsequent divergence as mechanisms to create natural variety and novel structures are now decade’s old theories (Ohno 1970). Basically, directed evolution approach is an application of the gene duplication concept. Gene duplication is seen as a way to avoid random sequences in evolution, because random sequences most often are not functional. Mutations in the duplicated genes explore the local sequence space and expand the number of members in a gene family.|
Pro-ID article? I think not, Dembski.
The article is a review of current protein-engineering processes, arguing out that while a rational, directed approach to protein engineering would be most efficient, at present we do not know enough to do this effectively and hence iterative searches of random peptides (referred to here as 'Darwinian' methods) are currently used (we have a program of this type running on my department's screensavers, apparently). The review states a number of drawbacks with this sort of method and argues that such methods are impractical in the long term. In addition, they argue that over-reliance on random search methods should not be allowed to impede progress towards a better understanding how to rationally design proteins.
No-where does this paper support the idea that evolution is flawed, although it does highlight some current areas of uncertainty, notably the question of how novel protein folds arise. No-one is pretending that uncertainties like this do not exist, and productive careers are currently being employed investigating these very problems. To suggest that this is a weakness is to misunderstand the very nature of science.
The review cites both Axe (2004) and Behe and Snoke (2004), which is probably why Dembski noticed it in the first place. I can see why he posted it; the abstract can be read in a pro-ID way (if you squint) and I would guess, oh, 0% of the pro-ID readers over there will actually take the time and effort to read the whole thing.