NIH: 100M Years to Change a Binding Site

[Frayed Knot]

Thanks.
So, was she the Neanderthal Eve as well?

I'm not sure whether anyone knows, at least not yet. The most recent ancestor to modern humans and Neanderthals was something like 200,000 years ago, so it seems doubtful that our LCA (latest common ancestor) was also theirs. But then it's made more complicated by recent work showing that there probably was some intermixing between sapiens and neanterthalensis much more recently than that.

 

[Frayed Knot]

So I'm gonna have to stick with the article and the radio show on this one. Information is defined by the reduction in uncertainty at the receiver's end. So the nature of the original signal is irrelevant.

Well, the information that we're talking about, with respect to the DNA genome, is the original signal of the communications example.

But here's where the distinction between Shannon information and layman's information becomes relevant. In order for evolution to proceed you need layman's information to increase.

Can you describe what this "layman's information" is exactly? If you say it must increase, does that mean that you can measure it? If you can measure it, that implies you have a definition of it. I know that Dembski tried to come up with something he called "Complex Specified Information," but he failed miserably.

My point here is that it's completely invalid to use the laws of thermodynamics or information theory to make claims about a property that no one can even define.

The only way to increase the Shannon information in DNA is to increase the genome length.

That's just false, as we've been discussing. Just like your typical English text has way less information that a random string of the same length, because of the fact that there is some predictability, every change to a string of data will have a small effect on the amount of information that it contains.

The only way to increase layman's information is for the author to write more code (we aren't good enough forgers yet ).

I'll evaluate the truth of this claim after someone defines "layman's information" and figures out a way to measure it. Keep in mind that mathematicians have been thinking about this for years and the result of their work is things like Shannon information and Kolmogorov information.

 

[fool]

I'm not sure whether anyone knows, at least not yet. The most recent ancestor to modern humans and Neanderthals was something like 200,000 years ago, so it seems doubtful that our LCA (latest common ancestor) was also theirs. But then it's made more complicated by recent work showing that there probably was some intermixing between sapiens and neanterthalensis much more recently than that.

I guess I'm confused. If we mixed are there people walking around with Neantertalensis Mitochondria?

 

[Frayed Knot]

I guess I'm confused. If we mixed are there people walking around with Neantertalensis Mitochondria?

Good question. I believe the DNA work done which supposedly showed intermixing was on the nuclear DNA and not mitochondrial. However, previous work along those lines was later shown to be contaminated with modern human DNA which led to erroneous results. This time they were supposed to be more careful, but it's fairly new data and is still pretty preliminary.

I don't know whether anyone has looked at mitochondrial DNA specifically looking for Neanderthal ancestors. If I get a chance, I'll ask around.

 

[Alate_One]

I don't know whether anyone has looked at mitochondrial DNA specifically looking for Neanderthal ancestors. If I get a chance, I'll ask around.

It's been looked into. In fact that was one of the first places that was looked at. It's a bit easier since there are far more copies of mitochondrial DNA per cell than nuclear DNA.

 

[fool]

It's been looked into. In fact that was one of the first places that was looked at. It's a bit easier since there are far more copies of mitochondrial DNA per cell than nuclear DNA.

And?

 

[Alate_One]

Your first point was that Bob didn't even know that the paper was against Behe, since you assumed he only read the abstract and not even the whole abstract.

You couldn't be more wrong. If you listened to the show instead of just reading the OP you would know that Bob and Fred were well aware that the paper was against Behe. I think you see a Real Science Friday thread and you feel obligated to come comment and in your haste to attack Bob you don't do your homework.

I'll readily admit I didn't listen to the show. Frankly, I've listened and watched plenty of creationist drivel down through the years and I've listened to Bob on occasion. But I'm not going to spend my life listening to EVERY *wrong* Science Friday show. And I don't need to, to know they are wrong the sound bites make that clear enough.

I will see if I can take some time this spring break and give this particular show a listen, as it might be entertaining. . . maybe. If so I'll report in a more detailed fashion.

Will you admit to being wrong about Bob not knowing the paper is against Behe? Careful, I might be making this up, you may want to listen to the show.

It doesn't really matter to the problems I've brought up with the general assertion that "math makes evolution impossible". It also doesn't make sense that Bob and co. are magically better at the math than the authors of the paper.

Nor has anyone yet ventured to answer the paper I posted earlier which renders the "information must increase" mantra rather pointless since it seems LOSS of "information" seems to have been key in human evolution. And it has a bearing on the current topic, which covers transcription factor binding sites. Again these apparently weren't changed, they were simply deleted in many cases.

Below is the Nature paper.

Cory Y. McLean, Philip L. Reno, Alex A. Pollen, Abraham I. Bassan, Terence D. Capellini, Catherine Guenther, Vahan B. Indjeian, Xinhong Lim, Douglas B. Menke, Bruce T. Schaar, et al. Human-specific loss of regulatory DNA and the evolution of human-specific traits. Nature, 471, 216-219 (9 March 2011)

Here is the science daily summary

edit- Also Stripe is in no position to be pointing fingers over not reading/watching extra-forum materials since he never does any sort of work of that kind. He expects everyone else to do it for him.

 

[Alate_One]

And?

No evidence of intermixing through observation of mtDNA. There was a recent paper that used statistics to detect intermixing in in nuclear DNA in certain human populations but not others. I'm always distrustful of things that can only be "discovered" by complex math, but I recall looking at the paper and it seemed reasonably solid.

There's another population of totally unknown humans recently discovered called Denisovans, and they appear to have interbred with Melanesians.

News story covering this here

 

[Stripe]

Well, the information that we're talking about, with respect to the DNA genome, is the original signal of the communications example.

Yip. And the only way to read that is to be on the receiving end of some sort of transmission.

Can you describe what this "layman's information" is exactly? My point here is that it's completely invalid to use the laws of thermodynamics or information theory to make claims about a property that no one can even define.

Information is any string of data deliberately arranged by an informed agent. It's not measurable with Shannon math which is why it needs to be distinguished.

That's just false, as we've been discussing. Just like your typical English text has way less information that a random string of the same length, because of the fact that there is some predictability, every change to a string of data will have a small effect on the amount of information that it contains.

If you're changing the code, you're transmitting. Shannon information is defined by what the receiver gets.

I'll evaluate the truth of this claim after someone defines "layman's information" and figures out a way to measure it. Keep in mind that mathematicians have been thinking about this for years and the result of their work is things like Shannon information and Kolmogorov information.

I know why it is so resistant to measurement.

I'll readily admit I didn't listen to the show.

Yet you're mocking Pastor Enyart and Mr. Williams for not having read what they are commenting on. You're a hypocrite.

edit- Also Stripe is in no position to be pointing fingers over not reading/watching extra-forum materials since he never does any sort of work of that kind. He expects everyone else to do it for him.

Show me one instance where I falsely claimed to have read something or insisted that others must read where I was unwilling.

You need to go away and have a good long think through your approach here. All your bluster is wearing thin.

 

[Flipper]

For me, the main difference is that Alate isn't presenting an Internet talk show in which she purports to have a clearer understanding of science than the overwhelming majority of professional scientists, before going on to scoff at both scientists and mathematicians for not sharing the presenters' fatally flawed and highly limited understanding of a complex topic.

 

[Alate_One]

Yet you're mocking Pastor Enyart and Mr. Williams for not having read what they are commenting on. You're a hypocrite.

The reason for my attack is not simply not having read it, but not having comprehended it. Then Enyart pretends to have a good handle on the issues by doing a talk show not only covering it, but asserting the paper *actually* says opposite of the conclusions that are clear even from the abstract. And THAT much is clear even from the quotes from the show.

Show me one instance where I falsely claimed to have read something or insisted that others must read where I was unwilling.

I didn't claim to have listened to Enyart's show. I have many times told you that the explanation on the forum of a complex issue was incomplete and you should reference a site, a video etc. You've always refused. This is the same thing. Pot, meet Kettle.

 

[Stripe]

For me, the main difference is that Alate isn't presenting an Internet talk show in which she purports to have a clearer understanding of science than the overwhelming majority of professional scientists, before going on to scoff at both scientists and mathematicians for not sharing the presenters' fatally flawed and highly limited understanding of a complex topic.

No. She's being a hypocrite.

I'll happily concede there are things from the show that might be wrong, but neither of you are prepared to discuss these things honestly.

The reason for my attack is not simply not having read it, but not having comprehended it.

:squint: Uh. OK. :idunno:

Whatever you say.

Then Enyart pretends to have a good handle on the issues by doing a talk show not only covering it, but asserting the paper *actually* says opposite of the conclusions that are clear even from the abstract.

Did he say something wrong? Did you even listen to the show?

And THAT much is clear even from the quotes from the show.

So we can totally ignore your screeching about how we should read the articles (or listen to the shows) before commenting.

I didn't claim to have listened to Enyart's show.

That you didn't. You said several things about the show as if you had listened to it and then attacked its presenters for having not read what they were presenting (when they did) making you a hypocrite because you had not listened to what you attacked.

It's perfectly simple, Alate. You're a hypocrite and a liar and your act is wearing thin.

I have many times told you that the explanation on the forum of a complex issue was incomplete and you should reference a site, a video etc. You've always refused.

Liar.

This is the same thing. Pot, meet Kettle.

It's not the same thing at all. If I deem a resource worth reading I will do so. I will not insist that others read something and then assert that I need not do the same.

That would be you.

 

[Alate_One]

It's perfectly simple, Alate. You're a hypocrite and a liar and your act is wearing thin.

And you are a total waste of time, Stripe. Your unsubstantiated liar accusations wore thin long ago. You refuse to substantiate them OR answer the substance of scientific data I've brought up in this thread. All you want to do is distract rather than discuss the issues at hand . . hmm what does that sound like? :think: Got that pot and kettle thing again.

It's not the same thing at all. If I deem a resource worth reading I will do so.

I didn't deem Bob's diatribe as worth a listen obviously. I know where Bob stands and I've listened to how he handles science before. The details of a scientific paper which are hard for anyone to get a handle on are on a different level than the details of a religious radio show segment that's run by a few non-expert, non-scientists.

I will not insist that others read something and then assert that I need not do the same.

Isn't that fun how that works? You will ask people to read your stuff, but you won't allow other people to ask you to read their stuff. I've read PLENTY of garbage you've posted in full and responded to it in detail but you seem to forget that. It's always the other guy that's got to do MORE work.

 

[Stripe]

And you are a total waste of time, Stripe.

Feel free to just ignore me. :idunno:

 

[Jukia]

It's not the same thing at all. If I deem a resource worth reading I will do so. I will not insist that others read something and then assert that I need not do the same.

.

Stripo: Last I knew you were an advocate for Walt Brown. That being the case your understanding of a "resource worth reading" is on a par with my understanding of Hungarian.

 

[Frayed Knot]

Yip. And the only way to read that is to be on the receiving end of some sort of transmission.

No, you're mixing up the concept of how much information there is to transmit, with how well the transmission channel can do its job.

We can read DNA perfectly. It seems like you're trying to equivocate around the question that's really at-hand. Our channel to receive the information from the DNA is not a limitation here - what we're talking about is the amount of information, the amount of entropy, in the DNA sequence itself. The more unpredictable it is, the more information that's contained in it.

Information is any string of data deliberately arranged by an informed agent. It's not measurable with Shannon math which is why it needs to be distinguished.

Is it measurable at all? If not, how can you make any claim about what has more information and what has less? If it is measurable, how?

"I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind."
- Lord Kelvin

If you're changing the code, you're transmitting. Shannon information is defined by what the receiver gets.

You're deliberately trying to obfuscate here. We're not talking about any transmission, we're talking about comparing two data sets and how much information each contains.

 

[Stripe]

No, you're mixing up the concept of how much information there is to transmit, with how well the transmission channel can do its job.

Not at all. The paper clearly states that information is not the same as entropy and that randomisation does not increase information. I'm trying to stick to that.

The way to do so is to recognise which end of the chain we are on. We do not determine what sequences we will see. We just read 'em.

We can read DNA perfectly. It seems like you're trying to equivocate around the question that's really at-hand. Our channel to receive the information from the DNA is not a limitation here - what we're talking about is the amount of information, the amount of entropy, in the DNA sequence itself. The more unpredictable it is, the more information that's contained in it.

If one letter change to another letter, we on the receiving end are not privvy to the change in conditions. The amount of information we might gain is unaffected. That's what the article says.

Is it measurable at all? If not, how can you make any claim about what has more information and what has less? If it is measurable, how?

It requires intelligent assessment on a case-by-case basis.

"I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind."
- Lord Kelvin

Shannon expressed information mathematically. We just do not pretend his model does the job completely.

You're deliberately trying to obfuscate here. We're not talking about any transmission, we're talking about comparing two data sets and how much information each contains.

The only way to define them as Shannon information is to assess the number of bits required to transmit them. And that number will most assuredly change with a change in the data set. But that is not how information is defined. Information is defined as the reduction in uncertainty at the receiver's end.

That's what the article (and the radio show) was saying and that's what I'm gonna go with until I hear something more convincing.

You gotta admit. The notion that randomness could increase information is pretty silly, right? Why would someone call randomness information?

 

[Frayed Knot]

Not at all. The paper clearly states that information is not the same as entropy and that randomisation does not increase information. I'm trying to stick to that.

That paper is talking about something different. It's confusing because a lot of the same words are used, but that doesn't mean that they're equivalent.

The paper is talking about randomization (noise) added by the transmission channel, it's not talking about the entropy of the original data set.

It requires intelligent assessment on a case-by-case basis.

So anyone who has this view can never claim that information is lost or gained, because there's no way to measure it.

You gotta admit. The notion that randomness could increase information is pretty silly, right? Why would someone call randomness information?

The idea that a less predictable set of data has more information than a more predictable set of data is accepted by every knowledgeable person. A less predictable data set would be one that has more characteristics of randomness.

You keep equivocating with Shannon's definitions of information, between the kind that's inherent in the data set, and the information transmitted by a channel. I'm starting to think that it's willful.

 

[Stripe]

That paper is talking about something different. It's confusing because a lot of the same words are used, but that doesn't mean that they're equivalent.

The paper is talking about randomization (noise) added by the transmission channel, it's not talking about the entropy of the original data set.

Well, perhaps you'd better go back and listen to the show with the correct use of the words in mind. :up:

So anyone who has this view can never claim that information is lost or gained, because there's no way to measure it.

:squint:

No. Anyone with a little smarts can tell easily the difference in value of information in two messages sent to them.

The idea that a less predictable set of data has more information than a more predictable set of data is accepted by every knowledgeable person. A less predictable data set would be one that has more characteristics of randomness.

Try reading the article again. :up:

Or perhaps you think Tom Schneider is not knowledgeable.

You keep equivocating with Shannon's definitions of information, between the kind that's inherent in the data set, and the information transmitted by a channel. I'm starting to think that it's willful.

Perhaps. "Information is always a measure of the decrease of uncertainty at a receiver". That's what I'm sticking with.

 

[Yorzhik]

If I have a batch of information that I want to convey to you, and the system that I'm using to send you the data substitutes some of its own randomness (noise), then the amount of my information that you are getting is less than what I started with. In communications systems, that's what's important.

But this is no different than a message being sent in a cell.