Re: increase in information represented by DNA

On Dec 13, 11:16 pm, Treus <treusd...@xxxxxxxxx> wrote:
John Vreeland wrote:
On Dec 13, 2:56 am, Treus <treusd...@xxxxxxxxx> wrote:
John Harshman wrote:
treusd...@xxxxxxxxx wrote:

John Harshman wrote:

Since information is the degree to which one variable of a system
depends on or is constrained by another, fitness is a function of
information inasmuch as one set of variables intrinsic to the organism
depends on or is constrained by another set of variables extrinsic to
the organism.

So genotype X has more information than genotype Y if genotype X has
higher fitness in the current environment?

Not necessarily. Information content is a function of the number of
elements in either corresponding set of variables. More complex
organisms require more information.

So that whole thing about measuring information by measuring fitness was
just wrong?

No, just incomplete.

While genomes clearly contain information, the concept is not usually
useful in the sense of being able to measure it. You cannot measure
information unless you understand it, and we are not quite there with
regard to genetics.

Which is why it is better to say fitness requires information, and
therefore the genome contains information by implication.

Perhaps so, but it seems a uselessly broad statement.

You'll know we really understand genetics when
scientists stop researching it. Imagine a child trying to estimate
the information content of a copy of _Science_. She may know the
alphabet as well as an adult, and she may have the entire issue open
before her, but it doesn't mean she will come up with any meaningful
answer to the question.

We have a hard enough time estimating the complexity of computer
files. One useful idea--which can be very effective if used properly--
is to compress the file and estimate its information content by the
size of the file. There are a couple of problems with this idea you
need to be wary of. One is that what you are really doing is
estimating the complexity of the original file plus the compression
utility. but if the latter is small then it can be ignored. The
second problem is that the utility has no way to identify random data
and will preserve it as information unless you tell it otherwise. A
stream of random digital noise can theoretically be compressed to a
single number representing the file size, but only if the compression
utility knows it is random, otherwise it looks like very dense
information and cannot be compressed at all. This is a rather
egregious case of "wrong."

Using a simple formula to estimate information will never work if it
contains some random data, which all genomes do. Of course, the
random data might one day be naturally selected. Does it become
information when it was selected or was it always information?

John Vreeland

It became information when was it was selected.

This had always been my opinion, but I am no longer certain. Simply
because a feature is not under selection pressure at the moment does
not mean it is not information. If the environment changes it may
come in and out of favor while organisms engage in cycles of
competitive arms races. When the American cheetah disappeared the
extreme speed of some of its former prey animals became something of a
redundancy, did the genes that coded for their speed cease being
information? At the moment they may be actively selected against due
to their need for resources that could otherwise be used for other
things, but if their environment changed the genes might be selected
for again. And the environment is always changing, even if this
amounts solely to relatively invisible competition between microbes
and immune systems.

Are genes that are no longer selected for no longer information? What
if they are broken pseudogenes? The fact that we can tell what a
pseudogene used to be used for indicates that it actually contains
information, even though it is not expressed as a protein in the
phenotype.

.

Relevant Pages

  • Re: increase in information represented by DNA
    ... information inasmuch as one set of variables intrinsic to the organism ... higher fitness in the current environment? ... single number representing the file size, but only if the compression ... contains some random data, which all genomes do. ...
    (talk.origins)
  • Re: increase in information represented by DNA
    ... information inasmuch as one set of variables intrinsic to the organism ... higher fitness in the current environment? ... single number representing the file size, but only if the compression ... contains some random data, which all genomes do. ...
    (talk.origins)
  • Re: Nando Explains Natural Selection to You!
    ... >> Thinking About Fitness and Natural Selection" ... > good eyed organism - doesn't reproduce ... beneficial* allele will go ...
    (talk.origins)
  • Re: Hamiltons rule in small population
    ... > group selection within evolutionary theory is as basic to evolutionary ... The controversy as to what exactly group selection is and if it ... It attempts to define organism groups as ... > centricity is it remains entirely heuristic. ...
    (sci.bio.evolution)
  • Re: Hamiltons Nonsense
    ... >> being other levels of selection than organism level. ... >> selection Hamilton assumed that his rule provided the ... > even cognizant of other levels is ludicrous - he was actively ...
    (sci.bio.evolution)