>I am curious Glenn. If we go through those 10^137 possible permutations at
>the rate of one a second, it will take 10^130 years. Suppose we have 10^100
>such molecules, it would still take 10^30 years. Do these numbers much
>larger than the age of the universe bother you? I recall reading somewhere
>that the astronomer Fred Hoyle believed that such large magnitude numbers
>would rule out randomness as a possible mechanism for change.
You don't have to attack the problem in the way that creationists want it
attacked. Selection is a powerful search algorithm. The note below is from
something I posted a long time ago. Note the probability of Joyce's ribozyme
that he is making. It is vastly more unlikely to find a RNA molecule with a
given functionality in this case than in the cytochrome c example.
The evidence that Yockey's number may be too small comes
from the relative recent developments in the techniques of
directed evolution. I first became aware of this evidence when I
heard of an article in Discover entitled "Speeding Through
Evolution," by Peter Radetsky, May, 1994, p. 83-87. Gerald F.
Joyce manufactured 10^13 random, 393-unit long ribozymes. He
put them in a vat with DNA and waited. He was looking for any
ribozymes which would perform the function of "DNA-cutting".
There are something of the order of 10^236 different permutations
of an RNA strand of this length. If only one permutation in the
entire phase space of this RNA was able to perform this function,
then one prediction would be certain. JOYCE WOULD NOT BE ABLE TO
FIND IT. Even if an analysis similar to Yockey's on cytochrome c
was able to be performed on this ribozyme and it was found that
10^120 different permutations would cut the DNA, Joyce still
would not be able to find any of them. He would have one chance
out of 10^116 of finding a functional DNA cutter. Thus under
standard views of the probability argument in relation to
biological polymers, Joyce should fail.
But he didn't fail!!! What he found was when he put his
RNA into a vat of DNA, 1 molecule out of a million was able to
cut the DNA. Originally all of them were very inefficient at it.
In the initial batch, it took an hour for the molecules to show
any activity. After an hour time he removed them from the vat,
analyzed which ones had cut the DNA, and set about reproducing
them in a mistake prone fashion. In other words, he allowed
these DNA cutters to produce mutated children. After two years
and 27 cycles of this, he had ribozymes which could perform the
task in 5 minutes.
Two things have a bearing on the probability argument.
First, 1 out of 10^6 molecules were able to perform the function
of DNA cutting! This is an incredibly large number considering
the usual treatment of this type of problem. Assuming that
Joyce's original ribozymes were truly random selections from the
entire phase space of permutation, implies that 10^230
permutations are able to perform the function, if inefficiently.
Since each batch of Joyce's ribozymes contained approximately
10^13 individuals, 27 cycles would produce 2.7 x 10^14 total
permutations. If you want an efficient cutter, then there are an
estimated 10^222 (10^236/10^14) of those - all with different
permutations. These original DNA cutters, while inefficient,
display one of the characteristics of points in a mathematical
phase space which are on the edge of an attractor. They point to
locations in the phase space where there is an attractor for the
functionality of DNA cutting. The real efficient cutters are in
close to the attractor. Secondly, the fact that Joyce was able
to find so many permutations, all different, which were able to
perform the proper function strongly implies that Yockey's
analysis is deficient in assuming that only the cytochrome c
family of permutations is able to perform the function of
Another piece of evidence that this is the nature of the
biological polymers is cited in a recent Scientific American
article. Joyce says,
"The target for DNA or RNA binding could be any type of
molecule, not just a protein. One of the first successful
selective amplification experiments was conducted in 1990 by
Andrew D. Ellington and Jack W. Szostak of Harvard Medical
School, who used small organic dyes as the target. They
screened 10^13 random-sequence RNAs and found molecules that
bound tightly and specifically to each of the dyes."
"Recently they repeated this experiment using random-
sequence DNAs and arrived at an entirely different set of dy-
binding molecules. When the selected DNA sequences were
transcribed into RNA molecules, they did not bind to their day
targets, which suggests that the DNA and RNA molecules bound to
the dyes by quite different mechanisms.
"That observation reveals an important truth about directed
evolution (and indeed, about evolution in general): the forms
selected are not necessarily the best answers to a problem in
some ideal sense, only the best answers to arise in the
evolutionary history of a particular macromolecule." - Gerald F.
Joyce, "Directed Molecular Evolution," Scientific American,
December 1992, p. 94
These are examples of why I think that the information was
already planned into the phase spaces of any given n-unit long
polymer. A proper analysis of the odds of finding a protein
which will perform a given function not only needs to take into
account the totally different permutations which occur in a
protein of length N but also all proteins of length N+1, N+2,
N+3... and N-1, N-2, N-3... which will also perform the desired
function. Only when this analysis is done can we truly say that
we know how likely or unlikely the origin of a particular
molecule is. In my opinion, the data coming available today
implies that the origin of complex molecules is not as unlikely
as has often been claimed. (See below for the effects this change
would have on the mathemathics presented by Yockey).
Joyce didn't have to search for a combination the way you suggested. And it
took him 2 years not the age of the universe!!!!!
Foundation, Fall and Flood