Re: "Genes are followers not leaders". Was: Birds of feather....

From: CNCabej (cncabej_at_aol.com)
Date: 12/24/04 

Date: Fri, 24 Dec 2004 22:56:52 +0000 (UTC)

N.C.
>> However, with a little research anyone could find such "crackpots" even
>> here in sbe. But my source is from a scientific work: "The direct
>specification
>> of an arbitrary wiring pattern of 10^4 connections between 10^10 neurons in
>the
>> human cerebral cortex would, for instance, take more than 10^15 bits of
>> information. This is to be compared to the little more than 10^9 bits in
>human
>> genome." (von der Malsburg, in The Handbook of Brain Theory and Neural
>> Networks, 2002). Now imagine: this immense information is only needed for
>> establishing neuronal connections in the cerebral cortex alone.

JM:
>Yes, but these 10^15 bits of post-development "information" might be encoded
>(i.e. compressed) in the 10^9 bits of the genome.

N.C.:
First, you see that the 10^15 bits is not my "crackpot". Second, from a
qualitative point of view, no one has been able to even speculate on how the
genetic information could provide that information.

JM:
> Or perhaps they are not
>really information at all - they are random connections. I suspect that both
>of these hypotheses play a big part in the explanation.

N.C:
They can be anything but random connections. On the contrary, in general they
are highly and amazingly specific.

JM:
A third hypothesis
>(yours?) -
>that a significant fraction of those 10^15 bits are carried by cytoplasmic
>factors in the egg - is obviously inadequate. The cytoplasmic factors don't
>carry anything close to the required amount of information.

N.C.:
To me this seems to be a misunderstanding: no one has ever hypothesized that
and certainly I didn't.

>> N.C. (previous posting):
>> The estimated number of maternal
>> cytoplasmic factors in a representative metazoan egg cell is ~10^4 ...
>[snip]
>
>JM:
>So I'm guessing that you realize that the amount of information that can
>be carried by these factors is much, much less than the amount of information
>that is carried in the genes.

N.C.:
Right, but that was neither claimed nor implied in my posting.

>> N.C. (previous posting):
>> There is clearly no genetic hypothesis on what determines gene imprinting
>in
>> the meaning that no genetic information has been shown (and by definition
>can't
>> be shown) to control the synthesis and functioning of methyltransferases.
>[snip]
>
>JM:
>I must be misunderstanding you again. Quite obviously (to me), if the
>methyltransferases are enzymes or ribozymes, then their synthesis and
>functioning is specified by genetic information. So, by "control", you
>must be talking about things like control of transcription or some kind
>of allosteric control of the methyltransferase operation. My reading of
>molecular biology is that these things generally involve a "switch" that
>gets "turned on" when there is a steric match between a signal molecule
>and something that was specified by the genome. For transcription control,
>there is a "promoter" sequence. For allosteric control, there is a domain
>of the enzyme. Both are specified by the genome.

N.C.:
There is no doubt that it is the genetic information that codes for the
synthesis of methyltransferases, but the real "control" in this case is that in
which transcription factors that you mention represent only downstream elements
of signal cascades that originate in the form of chemical signals in the CNS
(for some more detailed explanation see my reply to P.R. Jr in this newsgroup
on Dec. 23, 2004).

N.C.:
>I am not a professional molecular biologist - I am a control systems
>engineer. Any control system consists of a program or algorithm (designed
>and specified in advance) interacting with real time data (which appears
>during the operation) to generate a control response. I tend to think
>of the program as controlling, though I realize that there is an alternate
>viewpoint that sees the data as controlling. I'm guessing that you tend to
>see the data as controlling, and that is the reason why I am having such
>a difficult time understanding you.
>

N.C.:
I believe this is an advantage for you, for I believe any hereditary system in
living systems has to be a control system. Your thought-provoking idea makes
me think that the program in this control system in metazoans may be generated
via processing of the data from the developing metazoan structure rather than
be "programed in advance". However, I believe that after learning essentials of
my view, you may be able to elaborate on the nature of the epigenetic
information generated in the CNS.
I thank you again, ad even more convinced on the constructivity of your input.

Regards,

Nelson R. Cabej

Relevant Pages

  • Re: Quiet Airliners of the Future?
    ... >at least 25 minutes of electricity. ... would seem safe in guarding against 'Complete electrical failure', ... How many circuit breakers are involved in the Airbus flight control ... control system back into operation, ...
    (rec.aviation.piloting)
  • Re: Best Book on PID ??
    ... >of jargon in control system design that seems intended to preserve job ... PLL design with nonlinear tuning is ... >Lots of control systems have to work in situations where an ugly settling ... >where a poor transient response is that serious. ...
    (sci.electronics.misc)
  • Re: Best Book on PID ??
    ... >of jargon in control system design that seems intended to preserve job ... PLL design with nonlinear tuning is ... >Lots of control systems have to work in situations where an ugly settling ... >where a poor transient response is that serious. ...
    (sci.electronics.design)
  • Re: OT: Ladies and gentlemen, welcome to Hamburg, we will try landing again shortly.
    ... A robot could certainly do a better job of scraping the wing... ... the pilot did anything wrong it was to line up with the runway to early ... There are limits to what an automatic control system can ...
    (sci.electronics.design)
  • Re: Delphi version control
    ... >control system, START TODAY! ... >you're a single developer. ... I agree - EVERYONE should use some sort of revision control ... Third off: Proper build procedures. ...
    (comp.lang.pascal.delphi.misc)