Fossil discovery helps bridge gap between Ediacaran animals & those from the Cambrian

[rogue06]

Given you've posted in other threads since this, i'll take that as an assumption that you're giving up.

So, why are you so badly wrong? A random sample will differ at half its locations if and only if the objects being sampled are binary (ie, they can be represented as a zero or one). There are 20 amino acids and, last i checked, a set of 20 is not binary. Should they appear randomly, then the number of differences would be much, much higher.

Of course, the randomization in this case was done at the DNA level (4 of those, so not binary either), and amino acids aren't equally represented at the codon level (some are represented by six codons, others by just one). So the actual probabilities would be quite a bit more complex. But it's very, very clear that there will be more than 25 differences.

Congratulations, you've flunked both basic biology and basic statistics. You should really rethink the confidence you have that you understand biology well enough to argue against the best supported theory in the field.


With that out of the way, would you accept that i wouldn't make such a stupid error when i actually know better, and that my argument has nothing to do with the typical number of differences between a random sequence and a pre-specified one? Because it doesn't, and i've been saying it doesn't for pages now, and you appear to be completely unwilling to actually read either those disclaimers, or what my argument actually is.

I have read more than once that, due to the never ending barrage of challenges, that evolutionary theory is perhaps the most robustly attested theory in science in general. I think that there might be one or maybe two other contenders for that title but it probably would be a toss up between them. This one reason that its detractors are often left grumbling about it being "only a theory" displaying their ignorance about how theory is defined in science.

[Roy]

There are about a trillion malarial cells in an infected person, and this requires 10⁸ infected individuals to get to 10²⁰ generations.

Those aren't generations, they're organisms. By your reckoning, there have 7 billion generations of humans in the last century.

You literally have no idea what you're talking about.

[lee_merrill]

A random sample will differ at half its locations if and only if the objects being sampled are binary (ie, they can be represented as a zero or one). There are 20 amino acids and, last i checked, a set of 20 is not binary. Should they appear randomly, then the number of differences would be much, much higher.

No, at each position a residue is either different, or not. So it is binary.

... my argument has nothing to do with the typical number of differences between a random sequence and a pre-specified one? Because it doesn't, and i've been saying it doesn't for pages now, and you appear to be completely unwilling to actually read either those disclaimers, or what my argument actually is.

Well, you are disregarding my point here, which is why I repeat it.

Blessings,
Lee

[lee_merrill]

Those aren't generations, they're organisms. By your reckoning, there have 7 billion generations of humans in the last century.

Sorry, I used the term incorrectly.

Blessings,
Lee

[TheLurch]

No, at each position a residue is either different, or not. So it is binary.

In which case you just flunk math, because the rate of difference is going to be much higher, given that there's 20 amino acids.

Well, you are disregarding my point here, which is why I repeat it.

I'm disregarding your point because it has nothing to do with what i'm saying.

You could be talking about your fondness for Wallace and Grommit cartoons, and it would actually be more relevant, because it would at least avoid getting the math wrong. Though i would probably ignore it equally.

[lee_merrill]

In which case you just flunk math, because the rate of difference is going to be much higher, given that there's 20 amino acids.

No, each residue is either different, or it is not, there can't be 20 differences at one residue between a given sequence and a random one.

Blessings,
Lee

[TheLurch]

No, each residue is either different, or it is not, there can't be 20 differences at one residue between a given sequence and a random one.

No, but the chances of there being a difference go up from 1 in 2 all the way up to 19 in 20.

Think before you type.

[rogue06]

No, but the chances of there being a difference go up from 1 in 2 all the way up to 19 in 20.

Think before you type.

And here I thought that math was my weak subject. This is pretty basic stuff here.

[lee_merrill]

No, but the chances of there being a difference go up from 1 in 2 all the way up to 19 in 20.

Yes, the probability is greater, but the average stays the same, I would say.

Blessings,
Lee

[TheLurch]

Yes, the probability is greater, but the average stays the same, I would say.

That's the equivalent of saying "i have absolutely no clue about statistics."

[lee_merrill]

That's the equivalent of saying "i have absolutely no clue about statistics."

Why so? What do you think the average is?

Blessings,
Lee

[lee_merrill]

Well, it turns out I was wrong, the average increases with increasing base. I wrote a little program to simulate this, and the average for a sequence of 50 amino acids is about 47 differences.

import sys, random

sz = 50
loops = int(sys.argv[1])
sum = 0
for i in xrange(loops):
seq = [0] * sz
diffs = 0
for j in xrange(sz):
seq[j] = random.randrange(20)
if seq[j]:
diffs += 1
sum += diffs
 
print sum / loops

Blessings,
Lee

[TheLurch]

Well, it turns out I was wrong, the average increases with increasing base.

Wow, that's what, the second time you've admitted being wrong?

Also, you don't need a script. There are 20 amino acids. The chance at random of matching 1 of them is 1 in 20. 50 divided by 20 is 2.5. 50 minus 2.5 is, rounded off, 47.

It really is that simple, which is why Rogue and I have been stunned that you got it wrong.

[lee_merrill]

Wow, that's what, the second time you've admitted being wrong?

Also, you don't need a script. There are 20 amino acids. The chance at random of matching 1 of them is 1 in 20. 50 divided by 20 is 2.5. 50 minus 2.5 is, rounded off, 47.

It really is that simple, which is why Rogue and I have been stunned that you got it wrong.

Well, you all are smarter than I am, then. But I still hold that it is unreasonable to expect evolution to get 47 amino acids difference on average, to get to another interaction.

Blessings,
Lee

[TheLurch]

Well, you all are smarter than I am, then. But I still hold that it is unreasonable to expect evolution to get 47 amino acids difference on average, to get to another interaction.

Nobody's saying it had to. You're arguing against a straw man of your own making, not an argument anyone here has made.