Re: What are the odds?

On Nov 22, 11:40 am, Seanpit <sean...@xxxxxxxxx> wrote:
On Nov 21, 4:28 pm, hersheyh <hershe...@xxxxxxxxx> wrote:



[snip] 

*If* your model of evolution is that of the "747 in a tornado" straw
man, I agree that protein systems would not evolve and one can, *in
that case*, make a predictive probability calculation.  However, that
is not the model of evolution that scientists deal with.

Have any actual math or statistical analysis to back up
this little bald assertion.

In real evolutionary models, one cannot calculate probabilities based
on size alone, because in the real model of evolution, the gaps (in a
series of gaps) depend on the systems that exist in a particular
organism and not on total sequence space wandered randomly.  Your
statistics are GIGO.

In other words, if you know the gap size ahead of time, only then can
the ToE use real math to determine how long it would take to cross
it?

No. I am saying that doing a GIGO calculation like yours makes stupid
assumptions about how historical processes work. Like the assumption
that all actors have exactly the same probability of reaching some end
state. This argument works fine with a random lottery where, indeed,
every individual has exactly the same probability of winning. Even
there, of course, the more actors there are, the higher the
probability that *one* of them will win.

It would produce an irrelevant result if you used a method that takes
an average probability to estimate the probability that some country
will send a man to Mars in 20 years. The reason such an average
calculation is GIGO is because countries differ wrt their starting
points, their capacities, and their goals for the next 20 years. For
most of the 200 countries on the Earth, the probability that they will
launch a mission to Mars in the next 20 years is precisely zero or
close enough to it that the fact that it isn't zero is irrelevant.
There are only a handful of countries that, by a historical process,
have the capability to launch such a mission. The probability that
*some* country will launch a mission to Mars in the next 20 years is
better estimated by the probability that some country that already has
the capacity for manned space and a plan for doing so in the next 30
years or so will do it. That probability is not 1, but it certainly
is much higher (and more likely) than a probability calculated by
averaging in all those countries that have neither the capacity nor
the desire.

LOL - that's hilarious!  You do understand that the whole question
here concerns the ability to predict the gap size.  If you already
know the gap size, determining the average time needed to cross it
isn't a real prediction or science.  

And, because we are dealing with a situation which is not a random
lottery in which the probability of winning is independent of any
specific individual but the same for all of them, but rather a case in
which *some* organisms can have much, much smaller gaps than others (a
biased lottery), finding the individual cases where the gap is small
(by analysis of the most probable pathway indicated by historical
pathways) gives you a much better estimate (albeit usually after the
fact) of the gap that *really* needed to be crossed (the smallest
available one). Calculating gap size from total size is assuming that
evolution works like an unbiased completely random lottery. IOW, like
the "747 in a tornado" straw man version of evolution.

I'm asking you for a prediction
concerning what gap size is likely to be found at a given level of
functional complexity without knowing the actual gap size directly.

The *actual* gap size is the relevant gap size. The average gap size
under the assumption of a completely random lottery where no
individual is more likely to win than any other is not relevant
because history is rather important to the study of evolution. In
fact one might even say that evolution is a historical process where
the end result is a function of the past and not a random lottery
where past history doesn't matter and the probability of winning is
the same for everyone. But your understanding of evolution seems to
differ significantly from the idea that evolution is a historical
process.

You're predicting what the actual gap size is likely to be.

If you cannot do this, you're ToE is not based on real science.

The ToE *is* a real science. It just doesn't work by the method you
try to apply to it. The assumptions you make are invalid.

 It is
in fact based on your simple self-fulfilling prophecy that evolution
happens when it can happen. Only if your stories about how the gap
distances could be small are true could evolution happen.  

Which, of course, is why the presence of such reasonably well-
supported pre-existing conditions can be parsed out. The idea that a
new "function", say embryonic rather than adult beta globin, could
arise by duplication an divergence of a pre-existing beta globin to
fill that role requires that 1) there be a mechanism for duplication
and that the amount of divergence to differentiate adult and embryonic
beta globins not be onerously large. It doesn't envision producing
either one from scratch by a process that would be equally probable in
an organism that already had a beta globin gene and one (like amoeba)
that doesn't.

That's
obvious.  The question is, what are the odds that your stories are
true?  You don't know the answer to that question.  You don't have the
first clue.  Why not?  Because you don't subject them to any sort of
statistical analysis.  All you have is bald assertions and fantasy
just-so stories.  That's it.

I have repeatedly pointed out that there are statistical methods for
identifying that something occurred in a manner consistent with common
descent. But I am not interested in producing GIGO probabilities that
have no relevance to *real* evolutionary processes. I leave producing
such bull*** nonsensical irrelevant numerology to people like you.

Again, I keep asking you for some actual
statistical analysis and you keep coming back with bald assertions
that a few hundred million years can solve everything.  

The idea of common descent has *repeatedly* passed the statistical
requirement of consilience between changes in sequence into basically
an identical branching pattern. This despite the fact that different
genes change at different rates of change (due to the degree of
selection for retention of function). There is even consilience in
pathways based on morphology (where changes tend to be more often
selective) and sequence (where changes are largely neutral or near
neutral). Statistics overwhelmingly support the idea of common
descent.

Given that, are you proposing that your hypothetical *human-like*
intelligence that had a completely invisible presence on the earth
(talk about leaving a light footprint!) for hundreds of millions of
years worked by just interjecting a bacterial flagellar system here
and there occasionally?

Based on
what?  How do you know this?  Have any actual calculations or
estimates of the odds based on real statistical analysis of any kind
that actually deals with the mechanism of RM/NS?  

It certainly seems adequate to produce all the genomic differences
between chimpanzees and humans.  Hell, RM and neutral drift is
adequate to do that on an gross basis (there are about 50 or so places
where there has been change large enough to be likely due to selection
*for* change).  And there certainly have been some other differences
that due to selection for change, but without enough difference from
the rate of neutral drift to be identifiable as due to selection.
Small changes can have big effects.  Big changes (such as inversions
and chromosome fusions) can have small effects.

I'm not asking for an explanation of the similarities. I'm asking for
an explanation of novel functional differences.  

If the number of places where there is detectable sufficient
difference beyond that expected by chance drift alone is so small that
only 50 or so can be detected *even though* essentially the entire
sequence of both organisms is available, there either is no novel
functional difference between these two species (though they differ by
almost every bone in their bodies, not to mention things like brain
size, speech, hairlessness, etc.) or the differences that we observe
(excepting those 50 or so sites -- although the amount of change is
larger than drift can account for, none of these sites are so
different that RM/NS could not produce the differences) do not require
crossing large gaps. Take your choice.

Until you can
actually know the genetic basis for novel functional differences
(which fairly difficult to determine between humans and apes at this
point) you don't have anything to work with.  

Either there are no functional differences *you* would call "novel"
or, if they are, they clearly did not require crossing large gaps that
are impossible given the time available. The fact is that the amount
of difference could be due to chance alone. The probability that
there is some "novel functional difference" that required crossing
some large gap is essentially nil.

Why not deal with
something we know much more about? - like subcellular systems such as
the differences between the TTSS system and the flagellar system? - or
any other potentially beneficial subsystem in the proposed
evolutionary pathway?

Because the exact pathway by which they did form is, to some extent,
hidden in the distant past and the changes required blurred by the
more common neutral drift changes. That makes it hard to make
anything but reasonable guesses as to what some of the intermediate
gaps (note plural) were in the multi-step process producing the end
result. Besides, as I pointed out in an earlier thread, the *function*
of TTSS toxin transport *and* the *function* of motorized rotation
leading to organismal motility co-exist in the flagella of some
organisms. That is the gap between a TTSS function and flagellar
function can be and is, in this species, N-0 mutational events apart.
If I am not mistaken, there are mutations that knock out the flagellar
function without affecting the TTSS-like function of these flagella.
And, of course, in other bacteria, there are flagella that only
transport flagellar proteins and not toxins. The functions overlap.

No one in this forum or elsewhere in scientific literature actually
deals with the odds that their just-so stories are likely to be true.
Care to explain how you determine if what the theory needs to be true
is actually likely to have been true?  Any numbers here?  Any math?
What do you have besides the need for your stories to actually be
true?

I haven't seen a
single calculation from you explaining why 50 years isn't enough time
but a few hundred million years obviously is enough time.  Where is
your math?  Where is your science?

You yourself have said that changes in proteins of 300 aa that alter
function have been observed in this time frame.  How much harder is it
to alter function in a protein of 1000aa?  

You tell me!  How much harder is it?  Use real math this time.

I leave the GIGO bogus numerology math based on "747 in a tornado"
phony models of evolution to you. You tell me what the relationship
between size and ability to evolve is and the actual empirical
evidence you used to calculate it. You do have a series of proteins
of various sizes that you claim can and have evolved modified,
additional, emergent, or novel functions. Plot them against the gap
size that needed to be crossed to generate these functions and show
how the relationship can be extrapolated to larger sizes. I do
believe I have asked for such evidence before. Still haven't seen even
a list.

 And,
remember, we aren't just talking an alteration in the same type of
function here. We are talking about producing a novel function.

How much change in *function* is required to be considered "novel"?
Is beta hemoglobin a "novel" function unrelated to alpha hemoglobin?
Clearly, in mammals, you cannot substitute the two.

Of course, I would argue
that that depends on the *specific* starting state and *specific* end
state and not on size of the end system, which appears to pretend that
any new function must be built from scratch by random assembly or the
mathematical equivalent.

The question concerns the odds that a specific starting and ending
state are likely.  What are the odds Howard?  

What are the odds that some country will go to Mars in 20 years? And
how did you calculate that probability? What were the odds 20 years
ago that some country would have done it in 20 years (i.e., now)? How
did you calculate that number? Did you include all those countries
that lacked space programs in calculating the probability or did you
ignore them and just focus on those that had or have space programs?
Which number is more likely to be correct?

Where is your math
behind your bald non-statistics based assertions that the odds are
remotely "good" given any period of time to work with?  Have any real
calculations? I've yet to see a single formula or even an attempt at a
statistical estimate of the odds coming from you.  Still waiting . . .

I am definitely saying that the proposed mechanism of RM/NS is
clearly, given that the sequencing is almost complete, *more* than
sufficiently capable of producing all the observed genetic differences
that exist between humans and chimps. In fact, the vast majority of
differences that exist between these two species can be explained by
RM and drift. [BTW, that *is* saying that the *differences* between
these two species can be explained by known mechanisms of change that
can be experimentally observed. The *similarities* between the two
species, of course, is because they shared a relatively recent common
ancestor, or 'common source of similarity' if you prefer, as indicated
by many statistical tests of the pattern of similarity.]

But I will leave the GIGO calculation of probability as if evolution
were a non-historical random lottery to you.

Sean Pitmanwww.DetectingDesign.com

.