Re: Experimental basis for the Non-Beneficial Gap Problem

On Jul 12, 8:32 pm, John Harshman <jharshman.diespam...@xxxxxxxxxxx>
wrote:

Functional structure space is based on sequence space. That is why
the listed units of structure space here are in numbers of amino acid
residues. You can't get larger structures without more amino acid
building blocks.

This is all either wrong or gibberish, or both. The particular structure
space you directed my attention to has nothing to do with function or
sequence, but in fact solely with secondary structure: the raw data of
the similarity space is the positions of the alpha carbon in each amino
acid. Every amino acid has an alpha carbon. There is no information
about the identities of the amino acids in this data.

It is not the identities of the amino acids that is important - but
the absolute number per protein. The proteins analyzed were taken from
a database of real functional proteins. The illustration is "based
solely on objective similarity among the 3D structures represented by
C-alpha atoms and using a much larger structure database and
multidimensional scaling, revealed that all of the known protein folds
and protein structures cluster into four elongated regions in the very
sparsely populated protein structure space."

It seems to me then that the illustration is a 3D projection of
hyperdimensional protein structure space of real protein folds and
structures.

Here's the paper that describes the space itself:http://www.pnas.org/content/102/10/3651.abstract?ijkey=2462224295fc79...

(mind that nasty wrap)

Now the sizes that are color-coded in one of the figures are indeed
counts of amino acids. But the sizes have nothing to do with the
structure space itself, just the color coding in one figure.

Why use the 3D arrangement then - with different distances between
points in this arrangement?

The horizontal plane appears to represent
proportions of alpha-helix and beta-sheet present in the protein. Looks
like short proteins have a higher proportion of alpha helix than
proteins as a whole. This has nothing to do with anything you have said.

That's not all the illustration shows. It shows the relative
distances between proteins in structure space as a function of size as
well as the relative ratios of alpha, beta, alpha+beta, and alpha/
beta.

Yes. What it shows is that larger proteins are more likely to have A/B
structure, and that smaller proteins are more likely to have A only, B
only, or A+B structure. You will note that it also says that older
protein families are likely to have A/B structure than younger ones, and
infers that protein families tend to get longer sequences and to gain
A/B structure over time.

That's not all it shows. It shows that larger proteins that are of a
particular type, like All-Beta, are farther apart than smaller
proteins of the same all-beta type . . . etc.

As already noted for you, this concept is supported by the findings of
Dursten et.al., who actually proposes a mathematical formula to
estimate the "Functional Sequence Complexity (FSC)" of a protein
sequence. And, surprise surprise, there is a good correlation between
the size and sequence flexibility of a protein and Dursten's FSC
number. The greater the size and/or specificity of a protein, the
greater its FSC - as a general rule of thumb.

That might be true (though I doubt it), but it has nothing to do with
that diagram of sequence space you linked to. Given your gross
misinterpratation of the diagram and paper, I'm not all that inclined to
look up this new paper.

LOL - - whatever man . . . You can "doubt it" all you want. Why not
just read it?

http://www.tbiomed.com/content/download/figures/1742-4682-4-47-2.PDF

And by the way, the paper looks interesting. If you actually read it
rather than cherry-picking and misinterpreting figures, you might learn
something about how protein families evolve. Note that proteins, for
example, are not randomly distributed in structure space. Discuss why
this doesn't invalidate your estimates.

The beneficial sequences of any language system you consider are not
entirely randomly distributed in sequence/structure space. There is
always a certain degree of clustering that is present. What is
evident in all such functional/meaningful language/information systems
is that this clustering effect becomes less and less clustered at
higher and higher levels of functional complexity - the same is true
of functionally beneficial biosystems.

Yes, that's been your mantra for some time. But your big evidence in
this current round has nothing to do with either sequence or function,
or with beneficiality, or functional complexity, or any of your other
favorite buzzwords.

Read the Dursten paper - which I presented to you first and you
refused to read. The Choi/Kim paper was only used to show the
correlation of some stable folds/structures with size. You won't
actually read the other papers that actually deal with the concept of
functional complexity. What else can I do? I can only lead the horse
to water . . .

Sean Pitman
www.DetectingDesign.com

.

Relevant Pages

  • Re: Experimental basis for the Non-Beneficial Gap Problem
    ... the listed units of structure space here are in numbers of amino acid ... You can't get larger structures without more amino acid ... the absolute number per protein. ... Note that none of this is about size, sequence, or function, all of which you had claimed earlier. ...
    (talk.origins)
  • Re: Genetic sequences experiment
    ... So if sequence probabilities and genetics is your cup o tea, ... The protein functional code is nothing like words. ... amino acid sequence can have some function or multiple functions. ... it was removed and the bacteria had to evolve a new beta galactosidase ...
    (talk.origins)
  • Re: Genetic sequences experiment
    ... So if sequence probabilities and genetics is your cup o tea, ... The protein functional code is nothing like words. ... amino acid sequence can have some function or multiple functions. ... it was removed and the bacteria had to evolve a new beta galactosidase ...
    (talk.origins)
  • Re: Its A Matter of Time
    ... identical to that of chimpanzees if "the chimpanzee genome was not ... There was a technology that relied on protein ... we could sequence the gentic material itself. ... between chimps and humans had identical amino acid sequences ...
    (talk.origins)
  • Re: Its A Matter of Time
    ... identical to that of chimpanzees if "the chimpanzee genome was not ... There was a technology that relied on protein ... we could sequence the gentic material itself. ... between chimps and humans had identical amino acid sequences ...
    (talk.origins)