from Greg:==============================================
Brian Harper:
[...]
> >*acceptable* code. (Actually, this brings up an objection to the above:
> >if reverse transcription is accompanied by a random process which selects
> >codons within the space of possible sequences to code for a protein, then
> >all the information in the original DNA sequence can appear in the
> >reverse translated one.)
> >
>
> Before I forget, your protein that doesn't give a rip reminded me
> of Dawkin's selfish gene concept. Would this be a selfish protein?
:-) I suppose so. :-)
> But this process if it were to occur would seem highly destructive.
> Perhaps I'm goofing things up, but I had an intuitive feeling that
> the extra storage capacity of DNA might be a mandatory requirement
> for evolution to occur.
How so? My intuition isn't telling me anything here...
> Now onward. From the above and the subsequent interplay between
> yourself, Gene and others I realize that things may be getting beyond
> the point where I can make any meaningful comments [this is not
> to say I don't often blunder on in such situations anyway :)]
> due to my lack of knowledge in molecular biology. In fact, I'm
> tempted not to say anything for fear of interfering with the
> interesting discussion going on now. Alas, I can't seem to
> help myself :-).
Join the club. :-) I have my Micro textbook in front of me, though,
so I'm hoping to get meself eddycated here somehows.
There is a pretty good section that talks about 5' and 3' regions,
and introns/exons, and mRNA and tRNA and the whole thing, so I hope
to be more informed in a couple of days.
> Would it ever be correct to say that information theory prevents
> something from happenning? No, information theory is not a
> physical entity that it could get down and mingle with molecules
> and somehow influence their behavior. [as an experimentalist,
> I am under a contractual agreement to say something like this
> in any situation where theory encounters the real world ;-) ].
> Hopefully, all us realists agree here and apologies to any
> mathematicians whom I have offended.
Hear, hear! :-)
> Here I think we come back to a fundamental point, before
> we start talking about Shannon entropy of DNA sequences
> then we better be darn sure that the genetic information
> system satisfies the axioms of information theory. These
> axioms are not written on tablets of stone. There may be
> systems which convey information in some sense of the
> word but don't satisfy these axioms. But, once one is satisfied
> that the axioms are satisfied, then all the results of information
> theory follow perforce as a result of logic. So, I think what
> Yockey was saying is that the central dogma is not something
> peculiar to biology. It follows purely from logic (provided the
> axioms are satisfied) that it is impossible to transfer information
> from lower to higher entropy. Suppose we found a situation
> where this happens anyway. This would not imply that the
> impossible has occurred or that some mysterious or supernatural
> were occurring. What it would imply to me is that one had made
> a fundamental mistake and that one or more of the axioms
> of information theory doesn't apply. But if one concluded this
> then one would have to go full course with it. All talk about
> Shannon entropies of DNA in terms of information capacity
> become nonsense.
I see what you mean, and I think I've been attempting to say something
like that, although not so well. I think Shannon's assumptions
ARE met by DNA codon sequences, but that that may just not MEAN very
much in the actual world. i.e. it is certainly true that DNA chains
hold information, but just because one can maximize information content
on streams or write down expressions for mutual information doesn't
mean that these concepts refer to anything meaningful in biology.
> Now I want to return to your comments above. Other than the
> parenthetical comment I don't think there is any violation of
> information theory principles. Information is not really being passed
> to the source, instead, the information capacity of the source is being
> degraded until eventually its capacity is the same as that of the
> receiver. The parenthitical comment does however seem to go
> against the principles of info-theory as the DNA would be receiving
> more information than was sent. If I understood Gene correctly,
> he was giving some physical reasons for why this doesn't happen.
> This reminded me of something Yockey said in the quote I gave
> earlier:
I agree with this. I think to be much shorter about it, I was saying
that I didn't think the fact of degradation was the critical factor
to the Central Dogma.
> So, what I'm wondering here is whether we should consider
> the various things Gene has been saying as the physical
> justification for Yockey's statement "The genetic logic system
> has no such memory" or am I just clueless (I can't get over
> the hurdle of imagining how to receive more information
> than was sent, unless there is some memory effect or
> information is picked up in transit some how, but if the
> latter then where is it stored?)
No, I think you are right. In my new education, I am learning
that exons are regions of the DNA that are translated to RNA,
and then bonded together. i.e. in a degraded version of the pretty
picture on page 199 of my book, you have
<--exon---><-intron-><------exon----><--intron--->
and so forth, and only the exons are coded. That is, the introns
can have the cool inhibitors and stuff that Gene was talking about,
and Mr. DNA doesn't have to worry about coding for cyanide or
something accidentally (since those parts don't code).
After translation, you get <--exon---><------exon---->...
with all the introns gone, GONE, GONE!!! The point being that
any reverse translation process would have to take the sONE, GONE!!! The point being that
any reverse translation process would have to take the string of
exons, figure out where the introns had been, and what the introns
had been made of in the first place, and so forth.
Now another possibility strikes: if one could get this far, then
with my untrained imagination, I can envision enzymes that would
notice similarities of the exon string to the original DNA, grab
the new polypeptide, tear apart the exons, and insert the introns back into
(approximately) the right places. Of course there would be errors
introduced if the ends of the new exons had been changed. Gene? Is
this the raving of a molecular biology novitiate, or does a magical
enzyme like this exist in theory or practice?
-Greg