From: Martin Striz (mstriz@gmail.com)
Date: Mon May 15 2006 - 10:40:10 MDT
On 5/14/06, Mike Dougherty <msd001@gmail.com> wrote:
> Is the human genome a self-referential program code? Though I cannot seem
> to find a good example URL, I recall reading about genomics researchers
> identifying large sections of the the genome that do not appear to have any
> particular function.
After the publication of the final draft of the human genome, we know
that 1.2% of the sequence is protein-coding. Up to 20% may have some
influence of the product phenotype, for various reasons. 45% is
repetitive, meaning that our genome has been expanded (almost doubled)
by a bunch of parasitic genetic elements.
> It seems that the popular theory is that these
> sequences once provided useful information and that they are now obsolete
> yet have not been 'deleted.'
Probably the opposite. While we once thought that no non-coding DNA
is functional, we have discovered that microRNAs play an important
role in gene regulation. Do a Google Scholar or PubMed search on
microRNA.
> It would seem wasteful to leave the
> equivalence of developer's comments in the source code (unless we're
> currently running as un-optimized beta code)
The developer is evolution, and evolution doesn't care. Bad analogy.
> It is generally observed that
> Nature is highly efficient at managing energy/material resources.
Evolution is blind, so biological systems are a mosaic of highly
optimized and highly inefficient subsystems, all depending on
available selection pressures and past constraints. Exactly what we
would expect from a process without foresight.
> So what is the purpose of these apparently inactive sequences?
Like I said, up to 20% may have a role in gene regulation or something
else, then there's the fact that almost half is parasitic DNA. In
your software developer analogy, think of it as a second, third and
fourth programmer coming in and altering your code, and no
communication between you. In fact, everybody has a different goal
for the program in mind.
> Is it plausible that these "unknown format" sequences could be instructions
> that are parsed above the biomechanical level?
Some of them are parsed above the genomic level. Cf. alternative
splicing, protein folding chaperones, transcription/translation
regulation.
> Could there be a "rendering
> agent" that is acting on the aggregate of these sequences? Are we looking
> at the Bits of DNA looking for Bytes and missing the Words? Could these
> sequences be Turing rules for how to rewrite the tape containing the rules?
No, no, no and no. When you take the computer analogy too far, your
constrain your ability to understand the system. The genetic code is
/like/ a program to one extent or another, and carries information
kind of /like/ computer hardware does, but they are profoundly
different in some ways.
Martin
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:56 MDT