The Real Science Radio Caveman Show

Alate_One

Well-known member
Oops. Right you are. It should have been this:

ATAGGACCATAGA
TAGGACCATAGAA

The point stands. The same number of changes, vastly greater difference in similarity according to your brute force method.
It means you have two changes which are insertions and deletions. And no there isn't a vastly greater difference in similarity because the algorithm slides the two sequences so that the two changes stick out. It wants to optimize the number of matches and reduce the number of insertions and deletions. (You didn't play the game did you?)

Code:
ATAGGACCATAGA
 TAGGACCATAGAA


What does that matter? The sequence might still be present and a candidate for comparison.
It makes a difference to the organism. A frameshift is more likely to be deleterious than a substitution, thus it is scored with a penalty because it is more likely to be a FUNCTIONAL difference. Wasn't that your original point? About meaning? :dizzy:

The cell doesn't need to know. In a sequence of 109 randomly generated ATGCs, the (corrected) frameshift sequence I gave will show up about 50 times. If you are to pick out those and compare them to the original you will meet the problem I am describing.
No, you're comparing one extremely long set of ATGCs to another extremely long set of ATGCs. The longer the string the less likely you're matching by accident, which is how the algorithms work and scoring is done.

You can take two random strings and yes a handful of bases may match by random factors alone, but we're not talking about handfuls of bases matching. If you do the above you get a very low score and no homology.

Information cannot be reasonably compared using the brute force method you are describing.
Gee is that why there are literally thousands of papers that do this? But it doesn't work because Stripe says so? :rotfl:
(Then again I don't think you understood how the "brute force method" actually works)

Actually, I do. You might want to start elaborating instead of dropping off the kindergarten lesson and pretending that settles things.
You're at a kindergarten level at best.
 

Stripe

Teenage Adaptive Ninja Turtle
LIFETIME MEMBER
Hall of Fame
It means you have two changes which are insertions and deletions. And no there isn't a vastly greater difference in similarity because the algorithm slides the two sequences so that the two changes stick out. It wants to optimize the number of matches and reduce the number of insertions and deletions. (You didn't play the game did you?)
Which isn't making the comparison, it's preparing data.

Please try to stick with the issue we're discussing. :thumb:

Yes, sliding the sequence like in the flash game that I went through the tutorial on will make for much greater similarity. But the point remains. For the brute force comparison you insist is all that happens, the comparison between what is obviously very similar is much worse than the comparison between your first example.

It makes a difference to the organism. A frameshift is more likely to be deleterious than a substitution, thus it is scored with a penalty because it is more likely to be a FUNCTIONAL difference. Wasn't that your original point? About meaning? :dizzy:
Not if the sequence is a designed piece of code. That's possible, right? A properly functioning genome might see TAGGACCATAGAA as a short sequence within it, right?

No, you're comparing one extremely long set of ATGCs to another extremely long set of ATGCs. The longer the string the less likely you're matching by accident, which is how the algorithms work and scoring is done.
No, I'm comparing the very short sequences you first proposed and the corrected on I proposed.

You can take two random strings and yes a handful of bases may match by random factors alone, but we're not talking about handfuls of bases matching. If you do the above you get a very low score and no homology.
We're not trying to compare random strings. We're comparing the example pair you gave and the corrected pair I gave. Try to discuss the issue at hand instead of making up other things to talk about.

Gee is that why there are literally thousands of papers that do this? But it doesn't work because Stripe says so? :rotfl:
Who said it doesn't work? I said it was not useful. And you forget the charge you face. It is very obvious that much more goes into preparation of the data and the simple brute force comparison you maintain is all that happens is not all that happens.

(Then again I don't think you understood how the "brute force method" actually works)You're at a kindergarten level at best.
Great arguments. :thumb:
 

Yorzhik

Well-known member
LIFETIME MEMBER
Hall of Fame
You don't cherry pick "how they are compared".
Really? Is there a description of what method was used to compare these entire genomes?

BTW, the sizes of the genomes would matter a great deal in a "whole" genome comparison. But I'm sure that has to be mitigated in some way because functionality is what really matters. Do you know how they mitigated this problem in the study you cite?

You don't think the species on the list were a reasonable sample of placental mammal diversity?
It could be. Are there others available that were left out on purpose?

You can keep asserting this forever, but you don't have the information to say it is a problem. We have known that evolution happens and the earth is old from long before we had any idea of DNA data. Deal with it.
I'm dealing with it by being more unbiased than you are. When you said, "We're learning a lot about what those differences are. In the case of humans and chimps, most of them are in non-coding regions. And many (surprise surprise) are in regions associated with brain growth. There also appear to be a lot of rearrangements and duplications." It was really an admission that evolutionists haven't realized there might be a stepping stone problem and attacked it. I think they avoid admitting there might be a problem because they already know the answer means death to the main driver of evolution, which is mutation plus natural selection.

Maybe when they find a new plausible main driver to evolution they'll admit to a stepping stone problem. And I'll have to reassess my view of creation.

How about you admit that you'll never accept evolution, no matter what data is presented. You'll claim it's biased or move the goal posts every time.
That simply isn't true. I used to be a theistic evolutionist until I looked at the evidence. There was no pressure or reason to change except that evolution gets less and less viable the more we learn.
 

Frayed Knot

New member
Alate_one, give it up. Stripe is not interested in having an honest discussion. I tried to have a discussion about information theory last year, to no avail. Stripe comes along and throws up all over the thread, causing anyone interested in the discussion to go away.

That's how he sees his role here - Stripe doesn't contribute anything, but whenever someone comes along to present challenging ideas, Stripe's job is to make the discussion so unpleasant and unproductive that no one ever has to have their little bubble of beliefs challenged. He's the Official Threadsh*tter for this forum.
 

Alate_One

Well-known member
Which isn't making the comparison, it's preparing data.
Nope, that'd be the comparison. It's called an alignment what answers the question what is the maximum similarity between two sequences?

You use the same set of assumptions for everything you compare and you can get an idea of which sequences are the most similar to one another.

Here's a famous alignment. Maybe you remember this one? With the GULO genes.

Code:
Chimp –     CTACGCTTCACCTGGAG-GATGACATCCTACTGAGCCCC
Orang –     GTACGCTTCACCTGGAG-GATGACGTCCTACTGAGCCCC
HUMAN –     GTACGCTTCACCTGGAG-GATGACATCCTACTGAGCCCC
GUINEA –    GTGCGCTTCACCCGGGGGGACGACATCCTGCTGAGCCCC
RAT -       GTGCGCTTCACCCGAGGCGATGACATTCTGCTGAGCCCC

Now, which sets of organisms group together?

Please try to stick with the issue we're discussing. :thumb:
That would be you again. Please discuss actual data, not whatever your idea is of it. Thanks.

Yes, sliding the sequence like in the flash game that I went through the tutorial on will make for much greater similarity. But the point remains. For the brute force comparison you insist is all that happens, the comparison between what is obviously very similar is much worse than the comparison between your first example.

That's the method that is used to make the comparison, though normally done with computers.
 

Granite

New member
Hall of Fame
Alate_one, give it up. Stripe is not interested in having an honest discussion. I tried to have a discussion about information theory last year, to no avail. Stripe comes along and throws up all over the thread, causing anyone interested in the discussion to go away.

That's how he sees his role here - Stripe doesn't contribute anything, but whenever someone comes along to present challenging ideas, Stripe's job is to make the discussion so unpleasant and unproductive that no one ever has to have their little bubble of beliefs challenged. He's the Official Threadsh*tter for this forum.

Couldn't agree more, that's why he's been on my ignore list for years. He's a glorified troll.
 

GuySmiley

Well-known member
Alate_one, give it up. Stripe is not interested in having an honest discussion. I tried to have a discussion about information theory last year, to no avail. Stripe comes along and throws up all over the thread, causing anyone interested in the discussion to go away.

That's how he sees his role here - Stripe doesn't contribute anything, but whenever someone comes along to present challenging ideas, Stripe's job is to make the discussion so unpleasant and unproductive that no one ever has to have their little bubble of beliefs challenged. He's the Official Threadsh*tter for this forum.
And vice versa

Oh my, he can count! :chuckle:


You're at a kindergarten level at best.
Statements like these are so condusive to useful discussion don't you think?
 

Alate_One

Well-known member
Alate_one, give it up. Stripe is not interested in having an honest discussion. I tried to have a discussion about information theory last year, to no avail. Stripe comes along and throws up all over the thread, causing anyone interested in the discussion to go away.
Indeed, this should be obvious to anyone following this thread. Which is why he's going to get data to explain at every turn. And everyone can see his total failure at explaining any of it.
 

Stripe

Teenage Adaptive Ninja Turtle
LIFETIME MEMBER
Hall of Fame
Nope, that'd be the comparison. It's called an alignment what answers the question what is the maximum similarity between two sequences?
Aligning to find the best match is not the brute force comparison you described.

You use the same set of assumptions for everything you compare and you can get an idea of which sequences are the most similar to one another.
Of course.

Here's a famous alignment. Maybe you remember this one? With the GULO genes.
Code:
Chimp –     CTACGCTTCACCTGGAG-GATGACATCCTACTGAGCCCC
Orang –     GTACGCTTCACCTGGAG-GATGACGTCCTACTGAGCCCC
HUMAN –     GTACGCTTCACCTGGAG-GATGACATCCTACTGAGCCCC
GUINEA –    GTGCGCTTCACCCGGGGGGACGACATCCTGCTGAGCCCC
RAT -       GTGCGCTTCACCCGAGGCGATGACATTCTGCTGAGCCCC
Now, which sets of organisms group together?
The more similar the design, the more similar the genetics.
 

Stripe

Teenage Adaptive Ninja Turtle
LIFETIME MEMBER
Hall of Fame
Couldn't agree more, that's why he's been on my ignore list for years. He's a glorified troll.

Ah, it's Granite the fag. Ever the hypocrite and ever willing to expose himself as such.

Why not step out from behind your polystyrene wall and face those you're oh so willing to snipe at.

:loser:
 

The Barbarian

BANNED
Banned
Alate_one, give it up. Stripe is not interested in having an honest discussion. I tried to have a discussion about information theory last year, to no avail. Stripe comes along and throws up all over the thread, causing anyone interested in the discussion to go away.

But he makes a great example of YE creationist. My Mom used to say that no one was worthless; they could always serve as a bad example.
 

Alate_One

Well-known member
Aligning to find the best match is not the brute force comparison you described.
No, I did describe alignments and gave you an example from the start. You chose to make your own assumptions and resist correction.

The more similar the design, the more similar the genetics.
Is that why all the sequences I gave except the rat are non-functional?
 

Frayed Knot

New member
But he makes a great example of YE creationist. My Mom used to say that no one was worthless; they could always serve as a bad example.

Stripe:

mistakes.jpg
 

Stripe

Teenage Adaptive Ninja Turtle
LIFETIME MEMBER
Hall of Fame
Is that why all the sequences I gave except the rat are non-functional?

They probably aren't.

And you forgot to provide the answers to the important questions for these genes:
How is the base nucleotide sequence determined? Is it the sequence from an individual or is it a composite sequence? How is that composite composed? Are all the organisms sequenced in the same fashion?
 

Flipper

New member
They probably aren't.

And you forgot to provide the answers to the important questions for these genes:
How is the base nucleotide sequence determined? Is it the sequence from an individual or is it a composite sequence? How is that composite composed? Are all the organisms sequenced in the same fashion?

How about you try doing own legwork for once?

I know that for you it will involve the exhausting process of reading for comprehension and sounding out all the difficult big words and all the other unpleasantly brain-exerting intellectual exercises that you consistently shirk away from like a vampire at a garlic festival, but it's time to man up for once.

See if you can find some evidence, and then present it in the form of a counter argument. If you do a good job, I might even accept it as positive proof on the thread about your generally abject failure to be able to provide evidence in support of your assertions.
 

Stripe

Teenage Adaptive Ninja Turtle
LIFETIME MEMBER
Hall of Fame
How about you try doing own legwork for once?
How about you quit whining like a child for once? :allsmile:

Honestly - I have six-year-olds that go on less than you do.
 
Last edited:

The Barbarian

BANNED
Banned
(Flipper suggests that Stipe do his own research)

Stipe responds with his usual level of maturity:
How about you quit whining like a child for once?

He's asking you to man up and do your own research. And all you can do is throw a tantrum because he won't do it for you. You're perfect, Stipe.
 

Alate_One

Well-known member
Really? Is there a description of what method was used to compare these entire genomes?

BTW, the sizes of the genomes would matter a great deal in a "whole" genome comparison. But I'm sure that has to be mitigated in some way because functionality is what really matters. Do you know how they mitigated this problem in the study you cite?
There really isn't that much variation in mammalian genome size, especially compared to other organisms.

Very interesting talk on genomics and evolution is here:

Genome size



It could be. Are there others available that were left out on purpose?
Not that I'm aware of. There aren't that many mammalian genomes yet.

Maybe when they find a new plausible main driver to evolution they'll admit to a stepping stone problem. And I'll have to reassess my view of creation.
Mutation and natural selection are only SOME of the drivers of evolution.

That simply isn't true. I used to be a theistic evolutionist until I looked at the evidence. There was no pressure or reason to change except that evolution gets less and less viable the more we learn.
We find exactly the opposite actually. You should take a look at the talk I linked above. It shows patterns in genome size that make no sense from a creationist perspective but make a lot of sense from an evolutionary one.

For example, organisms that undergo complete metamophosis, typical frogs and salamanders have SMALLER genomes than those that stay in the larval state. Vertebrates that are stronger fliers have smaller genomes than those that are weaker fliers or flightless.
 
Top