lmc@denelcor.UUCP (Lyle McElhaney) (07/25/84)
I think I will try to show by example the mechanism by which chance is *not* brought to bear in a chemical (or biochecmical) reaction. I am paraphrasing Isaac Asimov in the essays I cited in my previous article. Let us examine the probability of assembling a water molecule from its constituent parts. A water molecule consists of an oxygen atom and two hydrogen atoms. If we simplify matters (I'll get into detail later), we can see that there are a number of ways in which the atoms could possibly unite to form *something*: O - H H-O-H O-H-H \ / 2 2 H 2 All other combinations are congruent with these three. The numbers are the number of possible ways that the three components could combine to produce that particular combination. In each case there are two possible ways, so using purely chance to define the ways in which these three molecules could combine, there is a 1/3 chance for each. This means (from a probabalistic point of view) that each form is equally likely, and that if 300 atoms of oxygen were combined with 600 hydrogen, there would be 100 molecules of each produced. Now, only one could be water; it is a tenet of physical chemistry that the form of a molecule is at least as important as its constituents in determining its properties. How is it, then, that when a mole of (atomic) oxygen and two moles of hydrogen are combined, we get exactly one mole of water? By the numbers we would expect 1/3 mole of water and 1/3 mole each of two other substances, but multiple experiments have shown that this doesn't happen. The reason is that the atoms have preferred modes of combining with other atoms to make up molecules. These are expressed in the structural notation of biochemistry by the number of dashs which link a given atom to its neighbors in schematic diagrams, like the ones I used above. Oxygen, for instance, has two such "bonds", while carbon has four and hydrogen one. These bonding numbers cannot be violated unless great stress is applied, and even then the molecules thus made generally break down when the force is eased. This being the case, it can be seen that there is in reality only one possible combination, the first displayed above, and therefore all H2O is indeed water. Further complications arise from the fact that physical chemistry takes place in three dimensions, and that the bonding counts are not the only forces which compel molecules to form as they do. Water, for instance has its two hydrogen atoms not exactly opposite each other, but rather they form an obtuse angle with respect to the oxygen atom in the center. Many of water's most interesting (and useful, not to say vital) characteristics stem from the asymmetry thus produced. These complications only make the du Nouey argument weaker, for rather than there being 3 possible combinations of H2O there are in actuality an infinity of possibilities, only one of which is really water. Enough. Bring on some better arguments, or quit the ring. -- Lyle McElhaney (hao,brl-bmd,nbires,csu-cs,scgvaxd)!denelcor!lmc
dann@bmcg.UUCP (07/25/84)
Regarding the discussion between Lyle McElhaney & Bob Brown:
If I remember my Scientific American articles in the area of
biochemistry, Lyle's argument about H2O does not apply here.
There are something like 20 different amino acids. Each of
them has a pair of chemical bonding areas, one on each side
of the molecule, which are identical from amino acid to
amino acid. Thus, the amino acids may be strung together
like beads with no particular preference as to the order of
stringing. In a soup of amino acids, with nothing guiding
the assembly order, one sequence is as good as another and
any results are purely random. Within a living cell, how-
ever, the order of protein assembly from amino acids is
carefully directed by DNA and RNA templates.
There is one added complication. Many proteins are "folded"
in complex fashions. Portions of the amino acid chains
overlap and are fastened together with weaker side bonds,
which do not use the two main connecting points on the
molecule. Within a living cell, this folding and fastening
is done by other complex proteins (called enzymes) which act
as catalysts. So, even if you accidently get the right
sequence of amino acids to form a particular protein, you
still have to fold and attach the side bonds. I think insu-
lin has only one side attachment.
Scientific American has published a book containing a col-
lection of articles describing this stuff. Really thought-
provoking, I highly recommend it to anyone interested.
By the way, insulin is really trivial by comparison to many
other vital proteins. Take a look at the structure of hemo-
globin sometime if you want to see a real work of art!
Dann McCreary
Burroughs Advanced Systems Group
{sdcsvax || ihnp4!sdcrdcf || cepu }!bmcg!dannlmc@denelcor.UUCP (Lyle McElhaney) (07/27/84)
Ah, you are correct. Alas, my example has only restricted application. However, there are two other points that need to be made to make the insufficient-time argument invalid (both have been touched on by others in the discussion; I merely reinterate): 1) The active site of the insulin molecule (in all of its various forms) is rather a small portion of the entire protein; in a general protein, it may be several short portions of the protein brought into juxtaposition by the three dimensional form of the molecule. This is how it is possible that pig insulin can be effective for humans -- the different parts can be radically different as long as the active site remains the same. The chances of putting together the relatively small active site randomly are much better than matching the entire protein. 2) The proteins were undoubtedly put together before the specific use for that protein became apparent. Thus, the first hemoglobin protein's oxygen binding site probably "fell into place" randomly before there was a requirement for it. Biological inertia kept it around until, one day, its ability to carry oxygen was "written into the script" of some organism, and from then on natural selection improved it's effeciency to do that one task. That hemoglobin happens to be the common oxygen carrying protein for all vertebrates does not mean that other mechanisms could not do it; it happened to be the one that was handy when the time was ripe. Please understand, I am not a biochemist, and therefore my arguments aren't as rigorous as they could be. My first argument was designed to show that the random sticky-marble theory of molecule/protein/dna creation is not appropriate because the atoms/amino acids have properties which encourage certain ways of binding. There are many other parts to the argument, which must all be considered as a whole to explain away the otherwise low probability of organic complexity. -- Lyle McElhaney (hao,brl-bmd,nbires,csu-cs,scgvaxd)!denelcor!lmc