Chimp genome again

[Montalban]

Sounds a bit Republican / Democrat to me.

The analogy means very little to me. Try using 'baseball' or 'ice hockey' or 'gridiron' metaphors

They'd have just as much meaning to me

 

[Mallon]

I was in error to say it was about behaviour. It doesn't specifically say this. What it does is it doesn't exclude it, by talking about all similarities

"The only reason chimps get the position as humanity's sister taxon among the apes is the repeated observation that they are the most similar of the apes to humans. It doesn't matter what the particular similarity actually is, chimps are the most similar to humans. "
Ibid.

I think maybe Todd had anatomical similarities in mind when he wrote that. Regardless, bonobo behaviour has been compared to that of humans, too:

Bonobo - Wikipedia, the free encyclopedia

The article goes on about saying the 99% similiarity in genes is irrelevant

"...because the actual percent identity (whatever that is) makes no difference at all. "
Ibid.

His point here is that the actual number (percent similarity) is irrelevant. What matters in inferring common ancestry is that our genes are more like that of chimps than any other animal. If our DNA was only 90% similar to that of chimps, and yet was still more similar to that of chimps than to any other animal, we would still infer exclusive common ancestry with chimps.

 

[Montalban]

I think maybe Todd had anatomical similarities in mind when he wrote that.

Yes, I understand. What I had done is inserted an emphasis not found there (though not excluded), based on the fact I know of such comparisons in behaviour.

However the behaviour comparisons are what I find interesting given the fact these two chimps behaviours are very much different from each other!

If we're comparable to two vastly different behaviour sets, I find that interesting

His point here is that the actual number (percent similarity) is irrelevant. What matters in inferring common ancestry is that our genes are more like that of chimps than any other animal. If our DNA was only 90% similar to that of chimps, and yet was still more similar to that of chimps than to any other animal, we would still infer exclusive common ancestry with chimps.

But that doesn't make sense either. The actual number is supposed to show that they're similar!

 

[sfs]

But that doesn't make sense either. The actual number is supposed to show that they're similar!

The point is that you can get many different answers as to how different we are from chimpanzees genetically, depending on exactly what question you ask. Whatever comparison you choose, however, you will find that chimpanzees (and bonobos) are more similar to humans genetically than is any other species.

 

[Montalban]

The point is that you can get many different answers as to how different we are from chimpanzees genetically, depending on exactly what question you ask. Whatever comparison you choose, however, you will find that chimpanzees (and bonobos) are more similar to humans genetically than is any other species.

SO if you measure them one way, they're close, but how close doesn't matter

 

[Mallon]

But that doesn't make sense either. The actual number is supposed to show that they're similar!

THAT we are related is evidenced by the pattern of similarity (nested hierarchy). HOW CLOSELY we are related is evidenced by the percentage of our DNA shared with chimps.

 

[Montalban]

THAT we are related is evidenced by the pattern of similarity (nested hierarchy). HOW CLOSELY we are related is evidenced by the percentage of our DNA shared with chimps.

That we are related is evidenced, in part by observation of behaviour, even though the two chimp types behaviour is so very different from each other.

How closely we are related - well that's supposed to be a point - chimps are our closest relative. That's comparing us to other species, and apparently chimps are our closest. But now the very comparison of 'how close' doesn't matter.

 

[shernren]

That we are related is evidenced, in part by observation of behaviour, even though the two chimp types behaviour is so very different from each other.

How closely we are related - well that's supposed to be a point - chimps are our closest relative. That's comparing us to other species, and apparently chimps are our closest. But now the very comparison of 'how close' doesn't matter.

Well, it's kind of like this. Let's say I administer a physics test to a class of 17-year-old students, where their answers are expected to be a sentence or two long to each question. Then I mark their questions. The brightest kid in the class gets 99%, and everybody else gets less than him.

Now instead of me marking their questions, their physics teacher marks their questions. Their teacher (I would hope!) knows more physics than me. So now there are more inaccuracies in the brightest kid's answer, and he gets 95%. But everybody else still gets less than him when the physics teacher marks all their papers too.

Now instead of their physics teacher marking their questions, we get Einstein, Feynman, and Hawking together on a panel and get them to mark the papers. Now everybody's an idiot, and the brightest kid only gets 50%. But surprise! Everybody else still gets less than him when their papers are remarked.

And finally we get a calligraphist in to mark the papers. The "brightest kid" is now the lowest in the class.

=========

Evolution makes predictions about the genetic sequence of related species. In particular, if we believe that a particular group of animals is related in a certain way - for example that chimps diverged from humans the most recently, followed by gorillas from the chimp-human common ancestor, and so on - then we must believe that their genetic similarities will be related in a certain way.

So it matters that chimps and humans are the closest species (if we believe that they diverged the most recently). Let's say we have one way of measuring their divergence that says they are 99.5% different. The next year, we invent a new way of measuring their divergence that says they are 95% different.

Does this prove that chimps and humans are less closely related? Well, that's like saying the brightest boy in the class stopped being the brightest boy in the class when he got marked by Einstein, Feynman and Dirac instead of by me. Clearly the brightest boy doesn't stop being the brightest boy simply because he is marked by different people. And clearly chimps don't stop being our closest relatives just because we're using one method instead of another.

That's because if any particular method of measuring divergence increased the measured divergence between chimps and humans, it would also (assuming the evolutionary hypotheses are true) increase the measured divergence between gorillas and humans, between macaques and humans, and so on. Within that pecking order of divergences, the chimp-human divergence would still be the least, and so the chimps would still be the closest related to humans.

What about observation of behavior? Unfortunately evolution predicts very little about that, because it only concerns genetic similarity. Two genetically very different creatures may have wings, because there may be two genetically very different ways to produce wings. Therefore the observation "dragonflies and dragons have wings" does not prove that dragonflies and dragons are genetically related. (For one, dragons are imaginary. We hope.) Similarly, bees and naked mole rats practice eusociality, again because there are genetically very different (or a very conserved genetic switch for) ways to make a eusocial species.

So there are some methods of measuring similarity that will not give you the right evolutionary relationships. However, since evolution is about inheritance, and genetics are inherited, evolution says a lot about genetics.

 

[Montalban]

Well, it's kind of like this. Let's say I administer a physics test to a class of 17-year-old students, where their answers are expected to be a sentence or two long to each question. Then I mark their questions. The brightest kid in the class gets 99%, and everybody else gets less than him.
...
So there are some methods of measuring similarity that will not give you the right evolutionary relationships. However, since evolution is about inheritance, and genetics are inherited, evolution says a lot about genetics.

No. People supposed that Chimps were closer relatives to us and then DNA testing is supposed to prove this, except it does and doesn't because people want to say that the degree of match is important (to show that we're related closely to them, than say to rats), but then it isn't important.

Your analogy is weird as if it supposes that answers in tests are only provisional answers... which is new to me. I always thought tests were to get results, not "We think you're right now, but later on we may discover with more research that you're not". Why they'd have you marking a test in a subject you're not good at is odd to me too.

You'd be like arguing then that an answer is more correct if it's given as 0.00000230034031 instead of 0.00000230034. If you want to do that I might as well argue (as James Gleick does in Chaos) that you can't actually measure the surface area of a table.

Evolution DOES look at behaviour. You've not looked at 'sexual selection' and the whole idea that the stronger lion get more mates. Check this site out. Darwin first talked of behaviour in animals and choices made in mate selection. Reproductive social behaviour is a part of evolution.

So there are some methods of measuring similarity that will not give you the right evolutionary relationships. However, since evolution is about inheritance, and genetics are inherited, evolution says a lot about genetics.

That's the hollowness of the whole comparison thing right there. We're 'close' but only on what you measure. So they predicted chimps would be our closest relations and based on choosing only some measures over others, they ARE!

At least that's what it seems to me.

 

[sfs]

SO if you measure them one way, they're close, but how close doesn't matter

If you measure them one way, we're closer to chimps than to any other species. If you measure them another way, we're still closer. As long as you're making the measurement genetically and it's at all a sensible measurement, you'll get the same answer.

How close is also interesting, but it's a different issue. That can give you an estimate of how long it's been since the human and chimpanzee lineages split.

 

[sfs]

No. People supposed that Chimps were closer relatives to us and then DNA testing is supposed to prove this, except it does and doesn't because people want to say that the degree of match is important (to show that we're related closely to them, than say to rats), but then it isn't important.

No, that's not how it went. Based on all kinds of physical similarities (including behavioral similarities), scientists inferred that humans were closely related to chimpanzees, bonobos and gorillas, but they really couldn't tell which species were the most closely related. Then they compared them genetically, and discovered that chimps and bonobos were most closely related, and then both of them to humans, and then all three to gorillas. This was new and interesting information -- that's what science is supposed to produce.

Evolution DOES look at behaviour. You've not looked at 'sexual selection' and the whole idea that the stronger lion get more mates. Check this site out. Darwin first talked of behaviour in animals and choices made in mate selection. Reproductive social behaviour is a part of evolution.

Sure, behavior affects evolution and is changed by evolution. But comparing behavior is a terrible way to determine evolutionary relationships -- which species are most closely related -- since it is not a quantitative measurement, and does not change at a steady rate.

There are some genetic comparisons you could make that would suffer from similar problems, but those aren't the ones we're making. Large swaths of the genome accumulate mutations at a pretty constant rate, and if you compare whole genomes, you will get an accurate measurement of the order and timing of relationships between the species.

 

[Montalban]

No, that's not how it went. Based on all kinds of physical similarities (including behavioral similarities), scientists inferred that humans were closely related to chimpanzees, bonobos and gorillas, but they really couldn't tell which species were the most closely related. Then they compared them genetically, and discovered that chimps and bonobos were most closely related, and then both of them to humans, and then all three to gorillas.

Bonobos weren't discovered until 1928. Darwinists waited until 1928 to do their comparisons?

This was new and interesting information -- that's what science is supposed to produce.

Well if science is about trivia, then so be it.

Sure, behavior affects evolution and is changed by evolution. But comparing behavior is a terrible way to determine evolutionary relationships

So Caroline Goodall wasted an awful lot of time then? Leaky should have sent her to go make some toast or something. And Fossey wouldn't have been killed!

This should read:
"Chimpanzees are efficient predators that use meat as a political and reproductive tool. Are there implications for the evolution of human behavior?"

"Apparently not!" - SFS

A very short article

-- which species are most closely related -- since it is not a quantitative measurement, and does not change at a steady rate.

It's it's not quantitative how do you know how steady it changes?

It must make Behaviorism a pseudo-science then

There are some genetic comparisons you could make that would suffer from similar problems, but those aren't the ones we're making. Large swaths of the genome accumulate mutations at a pretty constant rate, and if you compare whole genomes, you will get an accurate measurement of the order and timing of relationships between the species.

So if you measure what's similar you find what's similar?

 

[sfs]

Bonobos weren't discovered until 1928. Darwinists waited until 1928 to do their comparisons?

No, before "Darwinists" (usually better known as "scientists") knew about bonobos, they made their comparisons with the species they knew.

Well if science is about trivia, then so be it.

Science is about describing the natural world as it is and has been. If you consider that trivial, maybe you should find some other subject to discuss.

So Caroline Goodall wasted an awful lot of time then? Leaky should have sent her to go make some toast or something. And Fossey wouldn't have been killed!

This should read:
"Chimpanzees are efficient predators that use meat as a political and reproductive tool. Are there implications for the evolution of human behavior?"

"Apparently not!" - SFS

A very short article

Your response makes no sense. I said that behavior is bad for determining which species were most closely related. Were Jane Goodall (Caroline Goodall is an actress) and Dian Fossey trying to determine which species were most closely related to humans? If not, why did you bring them up?

It's it's not quantitative how do you know how steady it changes?

You can tell that it doesn't change at a constant rate by observing that sometimes it changes qualitatively in dramatic ways in a short time, and sometimes it changes hardly at all over long periods of time. Are you just trying to pick nits?

It must make Behaviorism a pseudo-science then

This statement is logically disconnected from anything else in this thread. Behaviorism (which is a philosophical position more than a scientific theory, as far as I can tell) says nothing at all about how useful behavior should be for determining evolutionary relationships between species.

So if you measure what's similar you find what's similar?

No. If we measure how similar things are, we find out how similar they are. If we compare objects' weights, we find out how similar their weights are. From that we can deduce what their masses are. If we compare species' genomes, we find out how similar their genomes are. From that, we can deduce how recently the species have diverged.

 

[Marshall Janzen]

I think that last paragraph in sfs' post is a good example for those who don't see why the different measurements for the same DNA comparison exist. If you tried to figure out which of three people weighed more, you'd get a different weight on earth than on the moon, yet the order (from heaviest to lightest) would always be the same.

Similarly, how the similarity between DNA is measured can lead to different results. Consider four genes in two organisms. One is identical while the others have minor changes. That would lead to a 25% similarity if measured at the gene level, or a drastically higher percentage if measured at the nucleotide level (where similarities within genes would count).

How similarity is measured at the nucleotide level is also more complex and can lead to different measurements. To give two examples, it depends on what algorithm is used to line up the genes to determine the best match, and it depends on whether a single mutation event that duplicates a range of nucleotides counts as a single deviation or one deviation for each nucleotide that was duplicated.

Use one methodology, and the numbers may be a bit higher than from a second. But, as long as both methodologies are well thought-through and are used consistently, the ordering of the results should be the same, even if the individual percentages come out differently.

 

[Montalban]

If you measure them one way, we're closer to chimps than to any other species. If you measure them another way, we're still closer. As long as you're making the measurement genetically and it's at all a sensible measurement, you'll get the same answer.

How close is also interesting, but it's a different issue. That can give you an estimate of how long it's been since the human and chimpanzee lineages split.

Aren't we closer to squids in some measurements?

 

[Montalban]

No, before "Darwinists" (usually better known as "scientists") knew about bonobos, they made their comparisons with the species they knew.

That's not what you said. You said that 'scientists compared gorillas, chimps ..."

Science is about describing the natural world as it is and has been. If you consider that trivial, maybe you should find some other subject to discuss.

You said you considered it trival, by having a task of 'coming up with interesting information'. It that's all, then that's rather trivial

Your response makes no sense. I said that behavior is bad for determining which species were most closely related.

Based on you saying so.

Behvaiourism measures behaviour, despite you thinking you can't do so.

Were Jane Goodall (Caroline Goodall is an actress) and Dian Fossey trying to determine which species were most closely related to humans? If not, why did you bring them up?

Sorry about the name. Yes, they were trying to observe similarities. That was my point.

You can tell that it doesn't change at a constant rate by observing that sometimes it changes qualitatively in dramatic ways in a short time, and sometimes it changes hardly at all over long periods of time. Are you just trying to pick nits?

I'm simply replying to what you write. You might believe something else but you keep trivialising science, and even dismissing beahviour science out-of-hand because of your (as yet unproven) claims that they can't do measurements of behaviour.

This statement is logically disconnected from anything else in this thread. Behaviorism (which is a philosophical position more than a scientific theory, as far as I can tell) says nothing at all about how useful behavior should be for determining evolutionary relationships between species.

It doesn't have to say anything about usefulness.

It believes behaviour evolves. And it measures change.

No. If we measure how similar things are, we find out how similar they are.

If you pubslish that things are similar based on only some of the things you measure, because they show similiarity, and ignore the others because they don't, then you get the trivial (but interesting) information you think science is about.

If we compare objects' weights, we find out how similar their weights are. From that we can deduce what their masses are. If we compare species' genomes, we find out how similar their genomes are. From that, we can deduce how recently the species have diverged.

Excepting that how closely they're related apparently doesn't matter

 

[Montalban]

I'm not continuing in this thread. I don't like people making unsubstantiated attacks on branches of science or claiming science is about 'interesting facts'.

 

[sfs]

Aren't we closer to squids in some measurements?

Not that I know of.

 

[sfs]

That's not what you said. You said that 'scientists compared gorillas, chimps ..."

Truly bizarre. When scientists knew about chimpanzees, gorillas and humans, they compared those species' morphology and behavior to determine which species were most closely related. When they learned about bonobos, they compared all four species. I made both statements because both statements are true; more than one statement can be true, you know.

You said you considered it trival, by having a task of 'coming up with interesting information'. It that's all, then that's rather trivial

I said nothing at all about it being trivial. All science does is come up with information, information that some of us consider interesting. That information includes things like how to blow up cities with hydrogen bombs, how to cure diseases, and how quarks bind. It's all information. You're the one who decided that meant it was trivial.

Based on you saying so.

No, based on the reasons that I gave you, and that you have not rebutted. Pretty much every biologist on the planet will tell you the same thing. It's hardly controversial that genetics provides better information about phylogenetic relationships than behavior.

Behvaiourism measures behaviour, despite you thinking you can't do so.

What behaviors has it measured that have been used to determine phylogenies? Please provide citations to the relevant papers.

Sorry about the name. Yes, they were trying to observe similarities. That was my point.

But they weren't using those similarities to measure relatedness, which was my point.

I'm simply replying to what you write. You might believe something else but you keep trivialising science, and even dismissing beahviour science out-of-hand because of your (as yet unproven) claims that they can't do measurements of behaviour.

I'm dismissing the study of behavior as a useful guide to phylogeny, simply because it isn't one. That in no way trivializes the study of behavior; it just means it isn't the right tool for that particular job. (Yes, some aspects of behavior can certainly be quantified. Has anyone ever done so in a way that was useful for determining phylogenetic relationships? If not, again, why are you bringing the subject up?)

It doesn't have to say anything about usefulness.

It believes behaviour evolves. And it measures change.

Please cite the behavioral studies that were used to measure evolutionary change.

If you pubslish that things are similar based on only some of the things you measure, because they show similiarity, and ignore the others because they don't, then you get the trivial (but interesting) information you think science is about.

Scientists really aren't as stupid as you suppose. We don't pick genetics as the tool to study evolutionary relationships because it happens to show a result we like. We use it because we know quite a lot about how genetic changes occur, and because we know that evolution is fundamentally about how genetic changes accumulate over time. Phenotypes -- body shape, size, behavior, and all the rest -- are reflections of the changes in genetics, and it is at the genetic level that evolution operates.

Excepting that how closely they're related apparently doesn't matter

It matters to those of us actually doing the studies.

 

[sfs]

I'm not continuing in this thread. I don't like people making unsubstantiated attacks on branches of science or claiming science is about 'interesting facts'.

As I recall, this thread was about the application of genetics to evolution. You were the one making an unsubstantiated attack on two branches of science (genetics and evolutionary biology), but implying that these studies were somehow biased or invalid for failing to compare behavior instead of genes. You have yet to support those suggestions with any scientific basis, and now you're going off in a huff when people challenge your claims. A pity.