The Once and Future Wallace

Real World Studies VI: Amino Acids.

     In 2009 I took a look at the bonding structure of amino acids to see whether this fit the model. Most amino acids have specifically four elements: carbon, hydrogen, oxygen, and nitrogen, and I was curious to see whether the bonding configurations among these might follow the architectural plan of the several other systems I've examined.

     Not being a chemist, I soon gave up on using bonding energy as the means of measuring interaction between atoms in each molecule, and opted for just a count of number of bonds for an initial try. Still, there seemed to be several possible ways of counting, so I looked at several. In my first try, I only looked at the better known amino acids that contained only those four atoms. The list: alanine, glycine, valine, leucine, isoleucine, serine, proline, threonine, phenylalanine, lysine, arginine, tryptophan, tyrosine, histidine, aspartic acid, glutamic acid, glutamine, and asparagine, eighteen in all.

     For alanine, C₃H₇NO₂, one of the resulting models for the matrix of interacting bonds is coded as follows (order of rows and columns: CNOH):

       10   1   3   4
         1   3   0   2
         3   0   4   1
         4   2   1   7

     This is then double-standardized as in the other examples given here, to determine whether a symmetric matrix of z scores is produced.

     For the seven different ways of counting, the following results were obtained:

Version 1b produced 11 that double-standardized to symmetry, and 7 that didn't.

Version 2a produced 16 that double-standardized to symmetry, and 2 that didn't.

Version 3a produced 14 that double-standardized to symmetry, and 4 that didn't.

Version 4a produced 16 that double-standardized to symmetry, and 2 that didn't.

Version 4b produced 13 that double-standardized to symmetry, and 5 that didn't.

Version 5a produced 6 that double-standardized to symmetry, and 12 that didn't.

Version 6b produced 5 that double-standardized to symmetry, and 13 that didn't.

     Thus, two of the models produced results that passed the test eight out of nine times. These are pretty impressive confirmations, considering the coarseness of the form of measurement.

     Encouraged, I used the three best counting methods to look at all the remaining amino acids listed in the Handbook of Chemistry and Physics that were made up only of C, O, H, and N. This added thirty-two more amino acids to the list. The results this time were not quite so good:

Version 1 produced 22 that double-standardized to symmetry, 9 that came close (were of the right general form, but not quite symmetric), and 1 that was not close.

Version 2 produced 25 that double-standardized to symmetry, and 7 that came close.

Version 3 produced 31 that double-standardized to symmetry, and 9 that came close (not adding to 32, because there were alternate forms of the measure that could not always be applied).

     Still, on the whole these are interesting results--remember, the initial simulations determined that, when double-standardized, randomly-populated 4 x 4 matrices with highest values along the diagonal only, produce a matrix of symmetric z scores less than two percent of the time. It remains for a real chemist to investigate the matter from a more exhaustive point of view.

_________________________

Continue to Next Essay
Return to Writings Menu
Return to Home

Copyright 2006-2014 by Charles H. Smith. All rights reserved.
Materials from this site, whole or in part, may not be reposted or otherwise reproduced for publication without the written consent of Charles H. Smith.

Feedback: charles.smith@wku.edu
http://people.wku.edu/charles.smith/once/amino.htm