Re: IC (Challenge)

Kevin O'Brien (
Thu, 18 Mar 1999 21:41:37 -0700

>>Here's a puzzle for ID supporters: can you solve it? Proteins contain
>>functional domains that must interact with each other in order for the
>>protein molecule to function properly. Remove even one domain and the
>>protein effectively looses its function. Would this mean that, generally
>>speaking, proteins are IC? If not, why not?
>They are, indeed, an example of IC.
>>Recently, however, the technology has now made it possible for biochemists
>>and molecular biologists to take different functional domains from
>>different proteins, combine them, and get functional proteins....
>Is this supposed to be an argument against intellegent design?

No, because the fact that these proteins are created using biotechnology has
no bearing on whether a chimeric protein would be expected to be functional.

>>The puzzle then is this: assuming that this proteinaceous chimera is IC,
>>how can you get an IC structure by assembling random non-functional parts?
>Were they random?

Yes. The basic techniques of the technology are by nature random; that
cannot be changed even if we wanted to change it.

>And if random, did you have a particular function in mind
>for the molecule produced within an intellgently designed enviroment?

No, and the molecule was not intelligently designed either. Random pieces
were randomly associated; they were not chosen for their function, nor were
they deliberately spliced together in any specific order. There was no
intelligent control over which pieces were recombined, or in what order they
were recombined, or in how they would interact if at all. The fact that
this is being done in a laboratory using biotechnology and not in a pond by
bacteria is irrelevant to the results.

>This is an old argument: "See, the floor can become the base of the
>moustrap." Something else can substitute for a part, but that something
>else performs the same function as the substituted part.

Except that with this technique you can be pretty sure that in most cases
the random piece DOES NOT perform the same function as the substituted part.
Yet you still get a functional protein in the end.

Remember, these parts are called **functional** domains. Proteins function
because their functional domains work cooperatively together. That's why
proteins loose their function even if just one domain is removed. For the
vast majority of simple proteins that perform a single function, the domains
of any one such protein do not differ radically in function from one
another. However, two proteins that perform radically different functions
will possess domains that function radically differently compared to either

Let's say I have two such radically different proteins: A and Z. The
domains of A all share pretty much the same function, as do all the domains
of Z, but because A and Z have radically different functions, the function
of a domain from A will have a radically different function from a domain in
Z, and vice versa. Let's say that I then remove a domain from A and a
domain from Z, then randomly assemble them together with domains from other
proteins. If all of these domains perform a radically different function
from all the rest, then you would not expect any of them to serve as
substitute parts with identical functions, as you suggest above, yet when
all is said and done, there is a good chance that the resulting protein will
have some kind of function. How can you explain this in light of IC?

>Not only that, someone with a purpose substitued the floor, or domain, for
>the original part.

If the process is truly random, which it is, there can in fact be no
guarantee that the domain will perform the same function as the original
part, and there is certainly no intelligence directing the placement the new

>>Or put another way, how can a functional structure that looses its
>>when even one piece is removed be made out of a random assortment of
>>non-functional pieces that were never "designed" to work together in the
>>first place, assuming that IC is a real concept?
>In this marvelous technology you described, you described design in the
>nature of the experiment.

But this design cannot influence the results. No one can use this technique
to purposely design a protein with a specific function. In fact you get
better results if you AVOID trying to design a protein with a specific

>Please give us more details to support your
>assertion that this experiment, or groups of experiments, had as part of
>their design, total randomness in how the molecules were rebuilt.

Fair enough. Better fasten your seat belt though; it's going to be a bumpy

First, select twelve simple proteins with functions as different as
possible. Then isolate their genes. Genes are made up of coding regions --
called exons -- divided by non-coding regions -- called introns. Each exon
codes for one functional domain. Using special enzymes, cut the genes up at
the introns so as to separate the complete exons. Add special sequences to
either end of each exon to make them "sticky" so that the exons will
recombine, then mix all the exons from all the genes together and allow them
to recombine. If the exons are thoroughly mixed, they will recombine
randomly, forming a wide variety of new "genes" of varying lengths and
varying compositions of exons in varying sequences. This technique is
called a chimeric library.

Separate the recombined exons on the basis of size, then use a technique
called PCR to create more copies of each recombination. Insert the
recombinations into circular pieces of DNA called plasmids, then insert the
plasmids into bacteria. The bacteria will make multiple copies of each
plasmid, but will also express the recombinations by making proteins from
them. Isolate these proteins and test them for activity.

Nothing could be simpler. Want to try it yourself?

>snip analogous argument with same question. Was it truly random, or did
>inherently intellegent behaviors affect the system?

I hope you're not suggesting that a scientist can, through sheer force of
will, force a random process to produce a specific, non-random result. The
only part that design actually plays in this process involves the bringing
together of certain random natural processes in a specific order, but the
randomness of the processes themselves insure that this kind of design
cannot influence the results.

Kevin L. O'Brien