Re: Pitman numerology

On Dec 9, 9:26 am, Vend <ven...@xxxxxxxxxxx> wrote:

Given a repeating DNA pattern, a frame shift mutation can indeed
produce a completely different repeating sequence using a different
residue in the sequence.

Alanine can be coded by GCT, GCC, GCA, GCG
Methionine can be coded only by ATG.
No shift can transform a repeating string of Alanine-coding codons
into a string of ATG.

There are other examples were it will work though. The codon TCT
codes for serine while the codon CTC codes for leucine. A sequence
that codes for a string of serine residues can experience a frame-
shift mutation that will produce a string of leucine resides.

 And, although large insertions and
substitutions must copy DNA from somewhere else, the copying process
itself, as far as the location to copy, is indeed arbitrary.  That is
why it can be classified as a truly "random" mutation.

It's random but the inserted or substituted sequences are not
equiprobable and are not sampled from a complete sequence space.

The odds that you will have the needed sequence within your genome
which, if inserted in a particular location would produce a higher-
level function is the problem here. The genomic library is very
limited. It doesn't have access to every sequence in sequence space -
as you point out. So, the odds that the needed sequence will actually
exist anywhere within the genome are what you need to consider here.
These odds get exponentially more and more remote with increasing size
and/or specificity threshold requirements for higher-level systems of
function.

That's the whole problem here . . .  There is an exponential drop in
the likelihood that a small mutation will be able to cross a linearly
increased Hamming Distance in sequence space.

A small mutation crosses a small mutation distance, by definition.
The Hamming distance in aminoacid sequence space crossed can range
from 0 to N.  What is the point?

I meant to say a small number of mutations.  The odds that a
particular Hamming distance between sequences A and B will be crossed
by one or two mutations drops exponentially as the Hamming distance
increases linearly.

Proof?

http://www.detectingdesign.com/flagellum.html#Calculation

Have any meaningful counter evidence?

Since we know that assuming that they
are uniformly distributed is unrealistic, we can't predict anything
without a realistic estimate of their distribution.

You can't assume anything else when it comes to finding targets with
unknown locations.  The best you can do is assume a uniform
distribution.

If the targets were uniformly distributed and scarce, we wouldn't
observe evolution.

That's right. We don't observe evolution when the targets are scarce,
beyond a certain ratio of targets vs. non-targets, and have unknown
locations. We do observe evolution only when the ratio of targets vs.
non-targets is fairly high - like when it is greater than 1e-30 or
so.

We observe evolution in action and historical data which is best
explained by evolution, therefore it's reasonable to assume that they
aren't uniformly distributed (or they aren't scarce).

They aren't scarce below the 200-300 fsaar level of functional
complexity - that's correct. They are very scarce beyond the 1000
fsaar of functional complexity. That's why we do in fact observe
evolution on the one hand, but not on the other.

The assumption is falsifiable.

Hardly . . . your own observations support it.

If you deny that assumption then the burden of falsifying it is on
you.

(Hint: get a large database of protein sequences, possibly with
corresponding DNA sequences.
Run conventional statistical tests for uniform distribution on them.)

Go ahead. Protein-based systems that require more than 1000 fsaar
show a pretty uniform distribution overall. Sure, there are clusters
of islands here and there, but all are not clustered into one tiny
corner of sequence space are linearly arranged in close contact like
your theory would need them to be in order for RM/NS to actually work
effectively. This isn't remotely true.

 Could the reality be that the actual location is closer
than the uniform distribution assumption?  Sure, but it could also be
farther away just as easily.

 Those odds
depend upon the ratio of targets vs. non-targets. . .  just as you
originally noted yourself.

I don't think I did.

You did mention the fact that you have to include a factor for the
likelihood of a certain number of mutations being able to do the job -
- did you not?

Please elaborate.

This is your own comment - you "elaborate" or explain what you meant
by it.

Sean Pitman
www.DetectingDesign.com

.

Relevant Pages

  • Re: Cantorian pseudomathematics
    ... >> And there's no particular question about what countable additivity ... >> sequence of pairwise disjoint events, ... The fact that each P_n is a uniform distribution is not relevant to ... assertion to another in the background, unnoticed, and uncared for? ...
    (sci.math)
  • Re: Real-Life Random-Number Generator
    ... > another basically random sequence with uniform distribution of the ... > sequence with unknown distribution and outputting a basically random ... last 2 or 3 digits of the e-count is random. ... Thus aliens in outer space, ...
    (sci.math)
  • Re: Real-Life Random-Number Generator
    ... > another basically random sequence with uniform distribution of the ... > sequence with unknown distribution and outputting a basically random ... Truck/SUV was second, ...
    (sci.math)
  • Re: Pitman numerology
    ... It would require an inserition or substitution of a large sequence. ... Given a repeating DNA pattern, a frame shift mutation can indeed ... Run conventional statistical tests for uniform distribution on them.) ...
    (talk.origins)