On Uncommon Descent, poster gpuccio has been discussing “functional information”. Most of gpuccio’s argument is a conventional “islands of function” argument. Not being very knowledgeable about biochemistry, I’ll happily leave that argument to others.
But I have been intrigued by gpuccio’s use of Functional Information, in particular gpuccio’s assertion that if we observe 500 bits of it, that this is a reliable indicator of Design, as here, about at the 11th sentence of point (a):
… the idea is that if we observe any object that exhibits complex functional information (for example, more than 500 bits of functional information ) for an explicitly defined function (whatever it is) we can safely infer design.
I wonder how this general method works. As far as I can see, it doesn’t work. There would be seem to be three possible ways of arguing for it, and in the end; two don’t work and one is just plain silly. Which of these is the basis for gpuccio’s statement? Let’s investigate …
A quick summary
Let me list the three ways, briefly.
(1) The first is the argument using William Dembski’s (2002) Law of Conservation of Complex Specified Information. I have argued (2007) that this is formulated in such a way as to compare apples to oranges, and thus is not able to reject normal evolutionary processes as explanations for the “complex” functional information. In any case, I see little sign that gpuccio is using the LCCSI.
(2) The second is the argument that the functional information indicates that only an extremely small fraction of genotypes have the desired function, and the rest are all alike in totally lacking any of this function. This would prevent natural selection from following any path of increasing fitness to the function, and the rareness of the genotypes that have nonzero function would prevent mutational processes from finding them. This is, as far as I can tell, gpuccio’s islands-of-function argument. If such cases can be found, then explaining them by natural evolutionary processes would indeed be difficult. That is gpuccio’s main argument, and I leave it to others to argue with its application in the cases where gpuccio uses it. I am concerned here, not with the islands-of-function argument itself, but with whether the design inference from 500 bits of functional information is generally valid.
We are asking here whether, in general, observation of more than 500 bits of functional information is “a reliable indicator of design”. And gpuccio’s definition of functional information is not confined to cases of islands of function, but also includes cases where there would be a path to along which function increases. In such cases, seeing 500 bits of functional information, we cannot conclude from this that it is extremely unlikely to have arisen by normal evolutionary processes. So the general rule that gpuccio gives fails, as it is not reliable.
(3) The third possibility is an additional condition that is added to the design inference. It simply declares that unless the set of genotypes is effectively unreachable by normal evolutionary processes, we don’t call the pattern “complex functional information”. It does not simply define “complex functional information” as a case where we can define a level of function that makes probability of the set less than 
. That additional condition allows us to safely conclude that normal evolutionary forces can be dismissed — by definition. But it leaves the reader to do the heavy lifting, as the reader has to determine that the set of genotypes has an extremely low probability of being reached. And once they have done that, they will find that the additional step of concluding that the genotypes have “complex functional information” adds nothing to our knowledge. CFI becomes a useless add-on that sounds deep and mysterious but actually tells you nothing except what you already know. So CFI becomes useless. And there seems to be some indication that gpuccio does use this additional condition.
Let us go over these three possibilities in some detail. First, what is the connection of gpuccio’s “functional information” to Jack Szostak’s quantity of the same name?
Is gpuccio’s Functional Information the same as Szostak’s Functional Information?
gpuccio acknowledges that gpuccio’s definition of Functional Information is closely connected to Jack Szostak’s definition of it. gpuccio notes here:
Please, not[e] the definition of functional information as:
“the fraction of all possible configurations of the system that possess a degree of function >=
Ex.”which is identical to my definition, in particular my definition of functional information as the
upper tail of the observed function, that was so much criticized by DNA_Jock.
(I have corrected gpuccio’s typo of “not” to “note”, JF)
We shall see later that there may be some ways in which gpuccio’s definition
is modified from Szostak’s. Jack Szostak and his co-authors never attempted any use of his definition to infer Design. Nor did Leslie Orgel, whose Specified Information (in his 1973 book The Origins of Life) preceded Szostak’s. So the part about design inference must come from somewhere else.
gpuccio seems to be making one of three possible arguments;
Possibility #1 That there is some mathematical theorem that proves that ordinary evolutionary processes cannot result in an adaptation that has 500 bits of Functional Information.
Use of such a theorem was attempted by William Dembski, his Law of Conservation of Complex Specified Information, explained in Dembski’s book No Free Lunch: Why Specified Complexity Cannot Be Purchased without Intelligence (2001). But Dembski’s LCCSI theorem did not do what Dembski needed it to do. I have explained why in my own article on Dembski’s arguments (here). Dembski’s LCCSI changed the specification before and after evolutionary processes, and so he was comparing apples to oranges.
In any case, as far as I can see gpuccio has not attempted to derive gpuccio’s argument from Dembski’s, and gpuccio has not directly invoked the LCCSI, or provided a theorem to replace it. gpuccio said in a response to a comment of mine at TSZ,
Look, I will not enter the specifics of your criticism to Dembski. I agre with Dembski in most things, but not in all, and my arguments are however more focused on empirical science and in particular biology.
While thus disclaiming that the argument is Dembski’s, on the other hand gpuccio does associate the argument with Dembski here by saying that
Of course, Dembski, Abel, Durston and many others are the absolute references for any discussion about functional information. I think and hope that my ideas are absolutely derived from theirs. My only purpose is to detail some aspects of the problem.
and by saying elsewhere that
No generation of more than 500 bits has ever been observed to arise in a non design system (as you know, this is the fundamental idea in ID).
That figure being Dembski’s, this leaves it unclear whether gpuccio is or is not basing the argument on Dembski’s. But gpuccio does not directly invoke the LCCSI, or try to come up with some mathematical theorem that replaces it.
So possibility #1 can be safely ruled out.
Possibility #2. That the target region in the computation of Functional Information consists of all of the sequences that have nonzero function, while all other sequences have zero function. As there is no function elsewhere, natural selection for this function then cannot favor sequences closer and closer to the target region.
Such cases are possible, and usually gpuccio is talking about cases like this. But gpuccio does not require them in order to have Functional Information. gpuccio does not rule out that the region could be defined by a high level of function, with lower levels of function in sequences outside of the region, so that there could be paths allowing evolution to reach the target region of sequences.
An example in which gpuccio recognizes that lower levels of function can exist outside the target region is found here, where gpuccio is discussing natural and artificial selection:
Then you can ask: why have I spent a lot of time discussing how NS (and AS) can in some cases add some functional information to a sequence (see my posts #284, #285 and #287)
There is a very good reason for that, IMO.
I am arguing that:
1) It is possible for NS to add some functional information to a sequence, in a few very specific cases, but:
2) Those cases are extremely rare exceptions, with very specific features, and:
3) If we understand well what are the feature that allow, in those exceptional cases, those limited “successes” of NS, we can easily demonstrate that:
4) Because of those same features that allow the intervention of NS, those scenarios can never, never be steps to complex functional information.
Jack Szostak defined functional information by having us define a cutoff level of function to define a set of sequences that had function greater than that, without any condition that the other sequences had zero function. Neither did Durston. And as we’ve seen gpuccio associates his argument with theirs.
So this second possibility could not be the source of gpuccio’s general assertion about 500 bits of functional information being a reliable indicator of design, however much gpuccio concentrates on such cases.
Possibility #3. That there is an additional condition in gpuccio’s Functional Information, one that does not allow us to declare it to be present if there is a way for evolutionary processes to achieve that high a level of function. In short, if we see 500 bits of Szostak’s functional information, and if it can be put into the genome by natural evolutionary processes such as natural selection then for that reason we declare that it is not really Functional Information. If gpuccio is doing this, then gpuccio’s Functional Information is really a very different animal than Szostak’s functional information.
Is gpuccio doing that? gpuccio does associate his argument with William Dembski’s, at least in some of his statements. And William Dembski has defined his Complex Specified Information in this way, adding the condition that it is not really CSI unless it is sufficiently improbable that it be achieved by natural evolutionary forces (see my discussion of this here in the section on “Dembski’s revised CSI argument” that refer to Dembski’s statements here). And Dembski’s added condition renders use of his CSI a useless afterthought to the design inference.
gpuccio does seem to be making a similar condition. Dembski’s added condition comes in via the calculation of the “probability” of each genotype. In Szostak’s definition, the probabilities of sequences are simply their frequencies among all possible sequences, with each being counted equally. In Dembski’s CSI calculation, we are instead supposed to compute the probability of the sequence given all evolutionary processes, including natural selection.
gpuccio has a similar condition in the requirements for concluding that complex
functional information is present: We can see it at step (6) here:
If our conclusion is yes, we must still do one thing. We observe carefully the object and what we know of the system, and we ask if there is any known and credible algorithmic explanation of the sequence in that system. Usually, that is easily done by excluding regularity, which is easily done for functional specification. However, as in the particular case of functional proteins a special algorithm has been proposed, neo darwininism, which is intended to explain non regular functional sequences by a mix of chance and regularity, for this special case we must show that such an explanation is not credible, and that it is not supported by facts. That is a part which I have not yet discussed in detail here. The necessity part of the algorithm (NS) is not analyzed by dFSCI alone, but by other approaches and considerations. dFSCI is essential to evaluate the random part of the algorithm (RV). However, the short conclusion is that neo darwinism is not a known and credible algorithm which can explain the origin of even one protein superfamily. It is neither known nor credible. And I am not aware of any other algorithm ever proposed to explain (without design) the origin of functional, non regular sequences.
In other words, you, the user of the concept, are on your own. You have to rule out that natural selection (and other evolutionary processes) could reach the target sequences. And once you have ruled it out, you have no real need for the declaration that complex functional information is present.
I have gone on long enough. I conclude that the rule that observation of 500 bits of functional information is present allows us to conclude in favor of Design (or at any rate, to rule out normal evolutionary processes as the source of the adaptation) is simply nonexistent. Or if it does exist, it is as a useless add-on to an argument that draws that conclusion for some other reason, leaving the really hard work to the user.
Let’s end by asking gpuccio some questions:
1. Is your “functional information” the same as Szostak’s?
2. Or does it add the requirement that there be no function in sequences that
are outside of the target set?
3. Does it also require us to compute the probability that the sequence arises as a result of normal evolutionary processes?
Distinct functions can contribute to to a whole. Is the steering wheel in your car the thing responsible for the totality of what makes transportation by car possible? No, but it is playing one part in the function of a car.
Different functions can add up to another function that is not identical with any particular one of them, but neverthess does emerge as a result of their combination.
I’m still waiting for you to actually argue and not merely asssert that we can’t do that. I’m going to keep doing it until you act to persuade me otherwise. With arguments instead of unsupported declarations that I don’t understand FI.
Corneel,
The 500 bit rule, for example, is based on counting 500 bits of FI in an observed protein. That protein in several of the cases that gpuccio used is a piece of a larger function.
Rumraket,
I think there may be an argument that there are 1401 bits in the system with two genes performing two different functions.
DNA_Jock,
The assumption here is there a selectable path exists between 1 and 1000 kcat that the enzyme has a reasonable chance to stay on. If the island that has a workable kcat exists, is separate and is 500 bits away your SOL.
Yes, you think you can arrive at the FI for “transportation” by adding up the FI of various car parts. If only cars were the only means of transportation.
That’s not in dispute. Are you even paying attention?
What is the function that they all, working together, contribute to, that is different from each individual function that they perform? Once you’ve defined that function then you can decide on a minimum level of function for that function, then you can calculate FI for that function.
You can do as you like. So can gpuccio. 🙂
So you believe that the ancestral sequence had no degree of function at all for LDH but then evolved to have some degree of function for LDH? You really do believe in magic. So tell me again why you believe that LDH evolved from MDH ancestral sequence when it had no LDH affinity.
wrong
Well, it was a forlorn hope…
Billy, the starting enzyme does both, just it does MDH really, really badly. And “no-one cares”. The 1000 and 1 numbers were co-efficients, made up to reflect the fact that “no-one cares” (meaning there is virtually no selection in favor of increased MDH activity). At these truly atrocious levels of MDH activity, it is scarcely contributing to fitness at all. Any mutants that did MDH better would be promptly selected against, because they would do LDH worse.
But, when the gene gets duplicated, the fitness landscape suddenly changes.
And my scenario explicitly assumed that there was NOT a selectable path between LDH and MDH, but (unsurprsingly) you failed to understand that.
Finally, using FI as a measure of distance, as you do here, is incoherent. I have mentioned this before.
DNA_Jock,
Sorry for misunderstanding. Where did the selectable path come from? Gene duplication? If so then I think your claim of no selectable path is false.
ok, I obviously dropped the ball there. If the reference sequence has no function, then FI=0, not infinite, right?
The math comes out infinity, but Hazen & Szostak define the case where the function is absent as FI being undefined. Which is another way of saying you’re not supposed to calculate FI for absent functions. It’s not that you can’t plug in the number 0 in the formula, it just gives a meaningless result. For reasons entirely unclear to me, this has not been obvious to Mung.
Rumraket cares 🙂
Is this even true? I thought that even beneficial changes are usually lost. These folks claim they are usually retained.
I went back and checked again in the paper and in my previous posts, and I have to correct myself here. The most ancestral version resurrected in the experiment did in fact have some, but extremely low degree of LDH function. And I even mentioned that in a follow-up post to the one where I first referenced the paper. I completely misremembered that and you were correct to say I did in fact say that.
This is where the chain unhinges, because there’s nothing magical if about if if that had happened. Why would there be?
There is nothing that demands it has the function to start with. Random mutations can cause alterations to the enzyme so it can accept another substrate it couldn’t before.
Interestingly they argue in the paper that the initial level of LDH affinity could not have been subject to selection before the duplication, and only really got “selectable” after a particular mutation subsequent to the duplication
“The reconstructed AncMDH2, which represents the last common ancestor of the apicomplexan MDH and LDHs, is a highly active and specific MDH, preferring oxaloacetate over pyruvate by seven orders of magnitude (Figure 6). The activity of AncMDH2 towards pyruvate is barely detectable, requiring a high enzyme concentration to quantify. AncMDH2’s kcat for pyruvate is 0.07 s−1, with a Km of 20 mM, while the physiological concentration of pyruvate is estimated to be about three orders-of-magnitude lower (e.g., ∼50 μM in human erythrocytes [Garrett and Grisham, 2005], the Plasmodium host during its blood stage). Based on these kinetic parameters, each AncMDH2 reduces one pyruvate molecule per hour. While the enzyme can be forced to reduce pyruvate in vitro, this negligible activity is unlikely to have been subjected to selection in vivo. Therefore, the various specialization hypotheses, which require a promiscuous ancestor, are poor models for apicomplexan LDH evolution. Activity towards pyruvate increased by over seven orders of magnitude on the evolutionary lineage between AncMDH2 and AncLDH, indicating neofunctionalization.”
Ahh, so ATP synthase and some ubiquitylation-related protein just adds up to 1 more bit than either of them in isolation?
Curious. Let’s hear that argument.
Cars don’t have to be the only means of transportation to be able to add up the FI for the individual functions of a car and find the FI for that particular mode of transportation by that car.
The function they all work together to contribute to is [Transportation using that particular car], vs the particular subfunction [steering using a steering-wheel], for example.
Again, different components of the car add to the car’s functoin, but they have individual functions of their own which do not by themselves fully describe the function of the car. There are entirely sensible answers to the questions “what is the function of the car?” and “what is the function of the gearbox?” and “what is the function of the carburetor?”
I think this analogy can also work for sentences in books, and genes in organisms. I really don’t see any issue here.
In theory yes. What’s the problem?
Yeah but presumably we are arguing because you think what I’m doing is wrong. But you seem unable to really show what’s wrong about it. You ask some questions which I assume you think make it obvious where I go wrong, but it seems trivial to me to answer them. What am I missing here?
I was just using your very own logic to show why you were wrong. And you seem to be confused about whether or not some function can be defined even though it has not evolved yet. Yet you’re arguing that the function “conversion of pyruvate into lactate” was there waiting for evolution to stumble upon it even though it had not yet evolved, and, at least according to you, the ancestral sequences could not perform that function even if only poorly.
But by absent they mean absent in all sequence space, if I got it right we were exploring the situation where a certain function is still not present (threshold set to function=0) but there are certainly functional sequences in the ensemble
If it was “activity towards” then it was goal-oriented and teleological. I have no problem with that. 🙂
Wrong about what? When I said that I think we’re not supposed to calculate FI for absent functions? I think I’m still right about that.
This makes little sense. A function can be defined without having evolved, but what does that accomplish?
This is a strange formulation. I don’t believe I have argued that the function was there waiting for evolution to stumble upon it before it had evolved.
I would argue that the enzyme was capable of the function at a very low level, but the beneficial (in the sense of having a positive effect on reproductive success of the organism) level of activity had not yet evolved. Essentially the capacity of the enzyme to reduce pyruvate to lactate did exist, but it was invisible to selection until a particular mutation of very large effect had happened to one of the duplicates. When that mutation had occurred, over subsequent generations the degree of function evolved to much higher levels as now the function had become visible to selection.
Yeah I was wrong when I stated that earlier.
There’s absolutely no reason to think that.
You can make up anything you like and say anything you like, just like Bill and gpuccio. I don’t have to show that you’re wrong. It’s up to you to show that you are right, and so far all you’ve done to that end is make assertion after assertion.
Hazen and Szostak have shown us how to calculate FI and you’re not doing it the way they have defined it. So call it rumFI.
You seriously don’t see what’s wrong with your “intuition” that the more genes something has the more FI it will have, therefore something with more genes must have a higher FI than a function that has only one sequence that meets or exceeds the threshold of function? That doesn’t even follow from your additive FI conjecture.
Let’s say that we take the novel Atlas Shrugged and make up a function, let’s call it the Atlas Shrugged function and define the minimum threshold to be must match exactly the sequence of characters which make up the novel. Then calculate the FI.
The sequence would be very very long, and the number of different sequences of the same length would be astronomical. Yet only one sequence would match. Imagine the FI.
And of course, for your scenario with multiple genes, there’s nothing that says they cannot have many sequences that meet or exceed the threshold unless you just define them not to. In which case, TSS and good for you.
I tried to edit my post but it woudn’t let me.
Wrong about there being an INCREASE in FI. You were trying to achieve an INCREASE in FI from a staring point that was either undefined or infinite.
Rumraket,
Ubiquitin related proteins depend on ATP synthase so they are different then the example you suggested.
In what way?
I know all that, but it doesn’t address my point: If a few mutations radically change the substrate affinity of a duplicated protein, like in the apicomplexan lactate dehydrogenase example (which, yes, is an observed protein), the novel function can easily start of with a protein that has more than 500 bits of FI. All that other stuff you cough up about the FI of the “original” protein or the FI of the “larger function” is irrelevant because we are specifically discussing the FI of the novel function, i.e. converting pyruvate to lactate. Hence, the 500 bit limit rule fails for this function, and thus fails as a general rule.
Rumraket,
For the ATP consumed in their reactions.
Corneel,
Why do you think you cannot infer design here?
Because the method gpuccio uses requires that we can rule out normal evolutionary processes. But in this particular example all the required mutations (a gene duplication, a six-residue insertion and a handful of single-residue substitutions) are well within the power of random mutation and natural selection.
Then you’re wrong, because ATP synthase is not the only way to make ATP. There are other enzymes capable of substrate level phosphorylation. All you can say is that some ubiquitylation protein is dependent on ATP.
Budt what does any of this have to do with whether two enzymes exhibit some particular amount of FI?
Corneel,
Sure. The pieces you describe add up to potentially about 30 bits if FI. Not well within the power of RMNS but feasible.
Only if the process of making the protein required 500 bits of information. In this case you described, the process requires substantially less bits.
On the other hand how many bits do think are required to duplicate a gene?
Rumraket,
So you believe that if we knock out ATP synthase from a eukaryotic cell then the ubiquitin system will continue to function?
I think gpuccio says essentially the same thing. However, you are arguing that those few changes introduced 500 bits of FI, but have by no means shown that to be the case. You’re simply assuming it. As is Rumraket.
It will continue to function until it runs out of ATP, whereever that ATP comes from.
But I’m still waiting on you to explain what the hell the FI of two different protein coding genes have to do with whether one makes and the other uses ATP.
Rumraket,
I believe the calculation is different between functionally independent genes and functionally interdependent genes. Can I knock out one and the other still works? Can the cell still function if one is knocked out?
I believe I have. All the issues you’ve attempted to bring up have failed and I have been able to explain in a simple and straightforward way why that is.
Sure I am, we set a minimum threshold and calculate FI for it.
But they haven’t stated anywhere that I’ve read that you can’t add up the FI for many individual functions and get the total FI for an ensemble of functions that contribute to a whole.
At best you could say that I’m doing something they haven’t explicitly said we can or can’t do. Yet I’m doing it and I don’t see what the problem is.
I actually wouldn’t say that is necessarily true. The most significant determinant of FI is sequence length, so it is possible that a single very large gene can have higher FI than several individual genes which even in combination are still smaller than the big one.
I agree that relationship is not necessarily always the case for reasons just explained.
Yes, I’m with you.
I’m not sure I follow what you’re saying here, so let me try to get some clarification.
Are you saying that there could be individual sentences in the “Atlas Shrugged” book, which contribute to the overall function of the book, with functions of their own, and these individual sentences could have many equivalent or even superior functions to the versions actually found in the book, and because of this, I can’t add up the FI of the individual sentences and pretend this is always equivalent to just looking at the book as one long sequence, set the threshold so only the book as-is qualifies, and calculate FI?
You need to keep your eye on what the right hand is doing.
Well Bill says that LDH and MDH add up, not to 1400 bits of FI, but 1401 FI.
Why? Well he says there’s “an argument” for that. And now I’m trying to get him to explain why, and how the situation is different for ATP synthase and ubiquitylation. He says it is, beceause ubiquitylation depends on ATP. Why does that make it different than the MDH and LDH example, well who knows?
But I suppose you’re right, Bill is in the habit of just blurting out shit he hasn’t thought about so my guess is he’s just making shit up as we go along in his endless grasping at straws, and I should stop feeding him questions that he’s just going to pile more assertions on top of.
Weird. I came up with 1404 FI. This FI stuff is complicated.
Quite. Which is why doing what amounts to work across multiple blogs and threads means miscommunication is difficult.
Mung, why don’t you ask gpuccio why he does not collect his ideas up into a single paper and publish it? It can then be built upon by other ID scientists. And then we can reference it, and specific paragraphs in it. This is more complicated then it needs to be.
Complicated situations require rigour. And it’s ironic, no, that gpuppcio is getting his “peer review” from a group of people he refuses to interact with while the cheerleaders at UD simply don’t know what’s going on.
FWIW, Puccio in fact agrees with Rumraket in that one gets to add up FI for complex systems
So?
So he would disagree with you.
That doesn’t follow at all. Another logic fail. But would you like me to ask him?
Now you know how I feel. 🙂
Hazen and Szostak don’t calculate FI the way you do. But you’re right, there’s nothing to prevent you from making shit up as we go along and then grasping at straws to defend your made up rumFI. Same goes for Joe and his popFI.
gpuccio @ UD:
Sorry dazz.
The problem is what I’m saying makes perfect sense, and it seems the only argument you have is that I’m doing something with FI that Hazen & Szostak doesn’t do explicitly (but also don’t argue that we can’t do).
You have tried to give some reasons for why what I’m doing is “wrong”, but they don’t demonstrate anything wrong about it. I have answered all your “but what if…” type questions in straightforward and sensible ways.
Even Gpuccio agrees we can sum FI for different functions. You’re just flailing around with no real argument to make here. And mere declarations from Gpuccio or yourself that I don’t understand FI aren’t automatically statements of fact. One wonders what it even is Gpuccio thinks I don’t understand, since I’m really just using FI the way he does.
dazz,
These are pieces of a larger function not two distinct different functions.