IS DNA sequence data used in analysis "cherry picked"?

[tas8831]

A creationist said yes, but he ignores the evidence.

Creationist:

"You have to cherry-pick tiny pieces of the genome between different taxonomic groups to find these genetic similarities and ignore all the genetic differences which are much larger than the similarities. That is a statistical mathematical blunder."​

I will never stop being astounded by the confidence and projection of people that are not as well informed as they want others to believe them to be.

One refutation:


Initial sequence of the chimpanzee genome and comparison with the human genome
The Chimpanzee Sequencing and Analysis Consortium
Nature volume 437, pages69–87 (2005)

"Best reciprocal nucleotide-level alignments of the chimpanzee and human genomes cover ∼2.4 gigabases (Gb) of high-quality sequence, including 89 Mb from chromosome X and 7.5 Mb from chromosome Y."​

The human genome is in the neighborhood of 3 Gb.

I may not be a self-proclaimed math expert, but let's see...

2.4 Gb is about 80% of 3 Gb....

"You have to cherry-pick tiny pieces..."​

Is 80% "tiny pieces"?

Wait - there is more:

"We calculate the genome-wide nucleotide divergence between human and chimpanzee to be 1.23%, confirming recent results from more limited studies"​

What was the claim - ah yes:

"all the genetic differences which are much larger than the similarities"
​

Which is larger - 98.77% or 1.23%?
The world may never know.

Let us assume for the sake of discussion that the 20% or so of the genomes not compared in this initial analysis were 100% dissimilar.

Which is larger - 21.23%, or 78.77%

Creationists are such jokers!

 

[pitabread]

That claim was a particularly odd one to make (although that thread was full of odd claims).

I wonder if he was setting up an argument for a Tomkins style approach to genetic comparisons (i.e. the utterly wrong way of doing it). But I suppose we'll never know.

 

[Ponderous Curmudgeon]

A creationist said yes, but he ignores the evidence.

Creationist:

"You have to cherry-pick tiny pieces of the genome between different taxonomic groups to find these genetic similarities and ignore all the genetic differences which are much larger than the similarities. That is a statistical mathematical blunder."​

I will never stop being astounded by the confidence and projection of people that are not as well informed as they want others to believe them to be.

One refutation:


Initial sequence of the chimpanzee genome and comparison with the human genome
The Chimpanzee Sequencing and Analysis Consortium
Nature volume 437, pages69–87 (2005)

'Best reciprocal nucleotide-level alignments of the chimpanzee and human genomes cover ∼2.4 gigabases (Gb) of high-quality sequence, including 89 Mb from chromosome X and 7.5 Mb from chromosome Y."

The human genome is in the neighborhood of 3 Gb.

I may not be a self-proclaimed math expert, but let's see...

2.4 GB is about 80% of 3 Gb....

"You have to cherry-pick tiny pieces..."​

Is 80% "tiny pieces"?

Wait - there is more:

"We calculate the genome-wide nucleotide divergence between human and chimpanzee to be 1.23%, confirming recent results from more limited studies"

What was the claim - ah yes:

"all the genetic differences which are much larger than the similarities"
​

Which is larger - 98.77% or 1.23%?
The world may never know.

Let us assume for the sake of discussion that the 20% or the genomes not compared in this initial analysis were 100% dissimilar.

Which is larger - 21.23%, or 78.77%

Creationists are such jokers!

Didn't somebody just post this.

 

[SelfSim]

What I saw there, was a model chosen to prove an assumed posit, whilst ignoring almost the entirety of the available objective evidence.

Science doesn't have the luxury of ignoring available evidence as the basis of its models.
Its not about proving 'the truth' .. more like its about finding reliable explanations for what's already there.

 

[Ponderous Curmudgeon]

What I saw there, was a model chosen to prove an assumed posit, whilst ignoring almost the entirety of the available objective evidence.

Science doesn't have the luxury of ignoring available evidence as the basis of its models.
Its not about proving 'the truth' .. more like its about finding reliable explanations for what's already there.

No what you saw there was somebody demonstrating that the original comment was unfounded, unless you are referring to the original creationist posting. please clarify.

 

[SelfSim]

No what you saw there was somebody demonstrating that the original comment was unfounded, unless you are referring to the original creationist posting. please clarify.

Creationist belief based arguments always turn out as being based on the belief in the truth of posits, whereas Evolution is based on a huge quantity of observations and thereby, not beliefs in the existence of some fundamental Truth.
(Ie: I think we're in solid agreement there ..):

A creationist said yes, but he ignores the evidence.

 

[Ponderous Curmudgeon]

Creationist belief based arguments always turn out as being based on the belief in the truth of posits, whereas Evolution is based on a huge quantity of observations and thereby, not beliefs in the existence of some fundamental Truth.
(Ie: I think we're in solid agreement there ..):

Thank you for the clearification, I found your original post ambiguous.

 

[tas8831]

That claim was a particularly odd one to make (although that thread was full of odd claims).

I wonder if he was setting up an argument for a Tomkins style approach to genetic comparisons (i.e. the utterly wrong way of doing it). But I suppose we'll never know.

Could be - he seemed pretty behind the times.

 

[tas8831]

Didn't somebody just post this.

But see, those were "tiny" pieces... just from all over the genome which ended up constituting really, really big chunks..

 

[SelfSim]

I found how abiognesis was roped into commentaries about alleles was particularly odd.
I don't think I've ever seen 'allele' used in any legitimate Abiognesis hypotheses?

That the proposed 'math model' was based on modern day DNA template based replication and that such a mechanism is never just assumed to have always been in existence, surely would have made obvious that something was not quite 'true' in general, about that model(?) This never seemed to occur however.

 

[FrumiousBandersnatch]

A creationist said yes, but he ignores the evidence.

Creationist:

"You have to cherry-pick tiny pieces of the genome between different taxonomic groups to find these genetic similarities and ignore all the genetic differences which are much larger than the similarities. That is a statistical mathematical blunder."​

I will never stop being astounded by the confidence and projection of people that are not as well informed as they want others to believe them to be.

This was in response to me reminding him that he hadn't offered an explanation for the human & chimp ERV correspondences I told him about in #1224:

"For example, why do we share over 200 endogenous retrovirus insertions (ERV) for a certain retrovirus group at the exact same points in our genomes, vs single figure counts for non-shared (unique) ERVs of that virus group?​

If there are around 10 million possible insertion points, what are the odds of sharing over 200 identical insertions by chance?"​

To call this 'cherry-picking tiny pieces of the genome' completely misses the point. He makes a point of this - when I asked him whether he had evidence that the dinosaur ancestors of birds did not have the loops of Henle (a type of kidney tubule) that modern birds have, he posted a link to an article on bird kidneys... completely missing the point.

His post #2059, in response to being asked how species form, betrays a surprising lack of understanding of speciation:

"I'm not saying that a mutation can't create a variant that can't interbreed with the parent species. But that new variant may have difficulty finding a mate. You might get a situation where two different variants might not be able to physically breed because of some limitation."​

At least, it surprised me - no mention of populations at all (it also misses the point and fails to answer the question)... am I missing something?

 

[tas8831]

This was in response to me reminding him that he hadn't offered an explanation for the human & chimp ERV correspondences I told him about in #1224:

"For example, why do we share over 200 endogenous retrovirus insertions (ERV) for a certain retrovirus group at the exact same points in our genomes, vs single figure counts for non-shared (unique) ERVs of that virus group?​

If there are around 10 million possible insertion points, what are the odds of sharing over 200 identical insertions by chance?"​

To call this 'cherry-picking tiny pieces of the genome' completely misses the point. He makes a point of this - when I asked him whether he had evidence that the dinosaur ancestors of birds did not have the loops of Henle (a type of kidney tubule) that modern birds have, he posted a link to an article on bird kidneys... completely missing the point.

His post #2059, in response to being asked how species form, betrays a surprising lack of understanding of speciation:

"I'm not saying that a mutation can't create a variant that can't interbreed with the parent species. But that new variant may have difficulty finding a mate. You might get a situation where two different variants might not be able to physically breed because of some limitation."​

At least, it surprised me - no mention of populations at all (it also misses the point and fails to answer the question)... am I missing something?

Those are exactly the types of claims I expect - and have seen - from people for whom the phrase "when all you have is a hammer, everything looks like a nail" was invented for.
Another currently active creationist wrote something of similar substance on a different forum wherein he was arguing for 'directed mutations' and how transposons were 'directed' and their insertions are 'nonrandom.' I explained:

...One of the major duplication events was facilitated by the insertion of a LINE between 2 of the genes.
The most common LINE in mammals is L1, and it is able to recognize the hexanucleotide "TTAAAA" and use that to insert itself into a genome. In that sense, it is non-random, since it uses a specific DNA sequence.

Would you like to guess how frequently that sequence shows up in genomes? Just for kicks, I searched GENBANK for the sequence for human chromosome 3. It is about 200 million BP, and my browser kept crashing, so I only downloaded 20 MB of it ( quick back of the envelope calculation indicates that 20 MB = only about 10 million 'letters' representing nucleotides, or about 1/20 of the chromosome in question). Once it loaded, I did a simple search for TTAAAA........................... How many times do you think TTAAAA showed up?

22,679 times.

That is, there are potentially 22,679 insertion sites for the L1 LINE in about 1/20 of just 1 chromosome.
But sure, transposon insertion is totally 'non-random'....wrt fitness....​

He declared this was "irrelevant"... It is always irrelevant when it does not fit their script.

 

[FrumiousBandersnatch]

... It is always irrelevant when it does not fit their script.

Yes, it is plain denial - it doesn't matter what the expert consensus or the big picture is, it must all be wrong because of the way one small piece of evidence is interpreted.

I was initially prepared to hear something from AK that would suggest the need for a change of emphasis or additional mechanisms for macroevolution, the way Kimura's neutral mutation genetic drift proposal did. When that didn't happen, I thought it was possibly a case of focusing too closely on those two lab experiments and missing the wider context, not seeing the forest for the tree; but to deny all the evidence that evolutionary theory was devised to explain, and all the accumulated evidence that has subsequently supported it is simply not rational

 

[Guy Threepwood]

By cherry picking tiny chunks of code in one application and aligning them with tiny chunks of code in another- you might similarly arrive at high degree of similarity between two entirely different software applications.

But of course in hierarchical digital information systems like DNA or our own software, the architecture and interconnection of these 'modules' are critical to the code's overall function. A single bit of information may determine whether or not an entire subroutine is activated. So too in DNA, a mere handful of genes can determine more complex cortex development and significantly more folding - determining what is arguably the most significant difference in development between chimps and humans.

 

[tas8831]

By cherry picking

What an odd way of admitting that you did not read the OP!

tiny chunks of code in one application and aligning them with tiny chunks of code in another- you might similarly arrive at high degree of similarity between two entirely different software applications.

So you are going to misrepresent this from the word GO.
Not surprising.

Say - weren't you the creationist that claimed that one could just swap out the specific terminology in a molecular biology journal and it would be indistinguishable from a computer science one? What is that saying about when all you have is a hammer?

But this part is awesome:

So too in DNA, a mere handful of genes can determine more complex cortex development and significantly more folding -determining what is arguably the most significant difference in development between chimps and humans.

Thank you for unwittingly letting the air out of one of the major, recurring creationist "arguments" against evolution - that if evolution were true, there would need to be some huge number of beneficial mutations making new alleles and darn it, there just isn't enough time.

I love it when creationists switch sides without realizing it.

Oh, also - emphasis added:

"We calculate the genome-wide nucleotide divergence between human and chimpanzee to be 1.23%, confirming recent results from more limited studies"​

Looks like the correct 'cherries' were picked...