As a control, 14 random sequences were generated using a tool designed to generate
random sequences:
http://ca.expasy.org/cgi-bin/get-sprot-entry?RND21334
The length of each sequence was 500 amino acids. The sequences were then aligned using the
same parameters that were used for G3PD. Of the 500 positions only one met the required
criteria to be included in molecular knowledge calculations. The procedure was repeated
several times, and the results were consistent. The technique described here usually
filters 499 of the 500 columns in the alignment. Usually, only one column scores well. The
only position that scored well in one example was all glycine. So this control experiment
only creates 4 bits of information (with the genetic code).
Furthermore, when using the genetic code to calculate knowledge and
information, every single position contributes some molecular knowledge - the knowledge to
not terminate the growing peptide chain. Of the 64 DNA codons, 3 terminate the chain.
Thus, at each position only 61 codons can be observed when 64 are possible. This is a form
of both knowledge and information. This effect more than offsets the 4 bits artificially
created above.
Perhaps more importantly, the technique described in this chapter
filters out many positions that contain knowledge, and these positions are assigned 0 bits
even though they contain quite a bit of knowledge. So this procedures always
underestimates the true knowledge.
next: ATP Synthase
home: The Origin and Evolution of Life
|
