Junk DNA, High R, Pinnipeds, and the Multiverse

[Rumraket (Mikkel R.)]

@Giltil

This is why I said that Gpuccio’s methodology for detecting design is designed to produce true positive but cannot in principle avoid false negative.

It also can’t avoid false positives (a highly conserved sequence could have recently broken and become nonfunctional, therefore have zero FI). But that’s the least of his problems.

His method straight up doesn’t work. It doesn’t detect design. Even if it could detect high FI (it can’t estimate FI from conservation though as already established), that still wouldn’t allow you to infer design as high FI can evolve. In fact, high FI can’t be designed if it can’t also evolve. The exact problem design is postulated to solve because evolution can’t, design can’t solve either. If there is no navigable fitness surface you are forced to do brute force guessing, which isn’t design it’s just random guessing, and then design and evolution are both hopeless.

Disagree? Then “design” your way to a long random password without getting any feedback whether you’re getting closer to the password. You can’t. The feedback is the very thing that makes design (and evolution) possible. If natural selection or something analogous to it isn’t bringing the guesses closer to the solution, there is no way to design your way out of it.

[Paul_King]

@Giltil

Gpuccio’s claim is that conservation through deep time indicates high constraints that indicates high FI that indicates design.

This is not quite true. Gpuccio claims that “large” increases in FI indicate design. That means that he must be able to measure differences in FI. And that means he can’t just be estimating lower bounds on FI - he must be able to show that one sequence has less FI than the other and be able to come up with a good estimate of the difference.

(Of course he can’t, and he can’t show that his presumed increases in FI are incompatible with evolution either)

[John_Harshman]

@Giltil

Gpuccio’s claim is that conservation through deep time indicates high constraints that indicates high FI that indicates design.

What would you think of a method for detecting design that gave 90% false negatives? As for the rest, @Paul_King provides a good summary.

[Giltil (Gilbert Thill)]

@Paul_King

This is not quite true. Gpuccio claims that “large” increases in FI indicate design

Agree. Admittedly, I’ve been a bit floppy in my description of Gpuccio’s reasoning. But that doesn’t affect the point I tried to expose to @John_Harshman.

[Paul_King]

@Giltil

Agree. Admittedly, I’ve been a bit floppy in my description of Gpuccio’s reasoning. But that doesn’t affect the point I tried to expose to @John_Harshman

The rest of my post explained why it negated your point. Again, Gpuccio needs to produce decent estimates of the change in FI - which just isn’t possible if your assertion was correct.

[Giltil (Gilbert Thill)]

@John_Harshman

What would you think of a method for detecting design that gave 90% false negatives?

If the method was able to give 10% or even 1% of true positive but no false positive, I would think it would be a terrific method, for it would revolutionize our world. Imagine the impact if it could be demonstrated that 10% or even 1% of the proteins were the result of design !

[Giltil (Gilbert Thill)]

@Paul_King

The rest of my post explained why it negated your point. Again, Gpuccio needs to produce decent estimates of the change in FI - which just isn’t possible if your assertion was correct.

I’m not sure to follow you here. What is the assertion of mine you think is incorrect ?

[Gisteron]

@Giltil

Imagine the impact if it could be demonstrated that 10% or even 1% of the proteins were the result of design !

Well, setting aside that FI is not a synonym for design, and pretending like design itself could be detected by some means other than external documentation, and that such means with an accuracy as low as that were discovered… We’d end up finding that there is some design in the DNA of organisms.

I’m sure the religious folk and philosophers would have a field day with that across the board. I reckon some of the scientifically minded among them might want to find some way of accounting for it, too. Meanwhile the way DNA interacts with its environment remains the same, the higher order impacts it has are also the same. So it will have no implications for medicine or agriculture. If having a bunch of deep-thonk-ers excited qualifies as revolutionizing our world, then I suppose that’s what it’d do. But in terms of actual mid-term impact on our civilization, I struggle to foresee any.

What difference do you reckon it would make, and how?

[Paul_King]

@Giltil

I’m not sure to follow you here. What is the assertion of mine you think is incorrect ?

If you had looked at the earlier post - or could even remember what is being discussed you would know.

But it really is a simple point. To measure increases in FI Gpuccio must be able to measure the difference in FI. If the changes he identifies as adding FI add only a little or no FI then they cannot be large increases in FI.

[Giltil (Gilbert Thill)]

@Paul_King

But it really is a simple point. To measure increases in FI Gpuccio must be able to measure the difference in FI. If the changes he identifies as adding FI add only a little or no FI then they cannot be large increases in FI.

The point I was discussing is whether the claim by @John_Harshman below was true.

@John_Harshman

Either most of the genome is non-functional or Gpuccio’s measure of FI is useless

I really don’t see how your simple point above addresses at all the issue of whether John’s claim is true or not?

[Paul_King]

@Giltil

I really don’t see how your simple point above addresses at all the issue of whether John’s claim is true or not?

Gpuccio’s measure of FI is certainly useless for his argument if it can’t provide a decent measure of the increases he’s trying to detect. Indeed, if all it can do is set a lower bound then I don’t see that it’s much use for anything.

[Mercer (John Mercer)]

@Giltil

Agree. Admittedly, I’ve been a bit floppy in my description of Gpuccio’s reasoning.

Gpuccio’s reasoning itself is so floppy as to be useless.

[John_Harshman]

@Giltil

But that doesn’t affect the point I tried to expose to @John_Harshman.

Ah, but it does. It’s not just about false negatives, it’s about the inability to reliably estimate the quantity of FI in a sequence. Because his whole point is about the differences in FI at various times in the past, if 90% of the FI in the current human genome is undetectable, there’s no way to know how much of the past FI is also undetectable, and therefore no way to say whether it’s increased or the rate at which it increased during the Cambrian explosion, if at all.

@Giltil

Imagine the impact if it could be demonstrated that 10% or even 1% of the proteins were the result of design !

If. Of course his method doesn’t actually measure FI, and FI is not actually evidence of design. But you can dream.

[Paul_King]

@Mercer

Gpuccio’s reasoning itself is so floppy as to be useless.

Which is why Giltil needn’t worry about “finding” another serious problem with it. It doesn’t make it any worse.

[Giltil (Gilbert Thill)]

@John_Harshman

Ah, but it does. It’s not just about false negatives, it’s about the inability to reliably estimate the quantity of FI in a sequence. Because his whole point is about the differences in FI at various times in the past, if 90% of the FI in the current human genome is undetectable, there’s no way to know how much of the past FI is also undetectable, and therefore no way to say whether it’s increased or the rate at which it increased during the Cambrian explosion, if at all.

Sorry but you’ve lost me with your reasoning here

[Giltil (Gilbert Thill)]

@John_Harshman

If. Of course his method doesn’t actually measure FI, and FI is not actually evidence of design. But you can dream.

This is irrelevant to the issue of whether a method of design detection that would produce 90% of false negative would have any value.

[John_Harshman]

@Giltil

Sorry but you’ve lost me with your reasoning here

Further explanation would require yet another rehash of Gpuccio’s arguments. Why not just reread the prior threads on that subject?

@Giltil

This is irrelevant to the issue of whether a method of design detection that would produce 90% of false negative would have any value.

True in the abstract. But it is relevant to whether Gpuccio’s method is of any value and to whether it detects design. But there’s no reason to go any further with this. Been done.

[Giltil (Gilbert Thill)]

@John_Harshman

But there’s no reason to go any further with this. Been done.

Agree. And thanks for this interesting conversation.

[Tim]

@Giltil

This is irrelevant to the issue of whether a method of design detection that would produce 90% of false negative would have any value.

A point of order, but, as established in a previous thread, Gpuccio’s Hot Mess isn’t “a method of design detection” so much as a method of “‘design assumption’ and ‘design assertion’”.[1]

That thread established that there has been no strong connection between Conservation and “Functional Sequence Complexity”, and that no substantiation had been given of any connection between either and either Functional Information or Design.

That thread was also memorable for your airy assertion that:

@Giltil

… the probabilistic resources of the whole universe are insufficient to find a target exhibiting 500 bits of FI from an unrelated state.

This assertion being rendered all the more ludicrous by the fact that, AFAIK, nobody, and particularly not you, Gpuccio, or anybody else in the “not properly filled out or developed” (i.e. “vacuous”) ID Echo Chamber, has ever even measured “500 bits of FI”, let alone developed a reasonable comprehension of the circumstances that could, or could not, give rise to it.

This renders that assertion, like most of your assertions on that and this thread, as nothing more than an articulation of naked fideism to the ID cause.

I would have thought that, at this late stage, your (and @colewd’s) fideism to that cause was sufficiently well-established that it could be left unsaid – but you both seem determined to remind of it, and of its vacuity, at every opportunity.

Addendum:

Given that in that previous thread, I opened by asking:

@Tim

Where have these detection methods been validated, including measurement of their rates of false positives and false negatives?

I suppose I should feel positively about the fact that the error rates of Gpuccio’s ‘method’ are at least being discussed. I would however feel more positively if these error rates were being measured, instead of merely being asserted.

[Giltil (Gilbert Thill)]

@Rumraket

It also can’t avoid false positives (a highly conserved sequence could have recently broken and become nonfunctional, therefore have zero

No. Suffice to work only with functional protein sequences.
Imagine a given functional protein in cartilaginous fishes and big apes that became recently broken in the human lineage. In that case, you would take the big apes sequence as the probe, not the human one, to assess whether a big information jump occurred at the time of the vertebrate transition and whether the sequence then remained largely unchanged through deep time.