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?
Recruitment of a hedgehog Regulatory Circuit in Butterfly Eyespot Evolution
I told you: gpuccio avoids this discussion by insisting that every E3 ubiquitin-protein ligase has some unique specific function that cannot possibly have evolved from some other E3 ubiquitin-protein ligase acquiring a novel target. But you acknowledged that a family of proteins can and does perform multiple functions that are specific cases of some overarching function. You opened the door yourself.
BTW: I like your position better 🙂
Corneel,
Speculation does not equal demonstration.
Once you buy into this evolutionary inference then you try and jam all square pegs into all the round holes.
Without the design inference you cannot explain the modifications.
Corneel,
I don’t understand how one follows from the other. Most nuclear proteins have multiple targets but this does not equate to easier to evolve. This supports the design inference.
That makes zero logical sense Bill.
Fitness is a measure of reproductive success for all organisms. Their ability to produce more offspring than their competitors. It doesn’t matter how they achieve this. If mutations in ATP synthase cause lower fitness, then they are selected against even within the same population for the same species. There is no reason to do a cross-domain comparison.
Why are there no E coli cells with particular mutations in ATP synthase? Because they’re selected against. To make a statement like this we don’t need to look at eukaryotes at all. One E coli cell compared to another is all we need here. If E coli cell #1 has a mutation in it’s ATP synthase gene that E coli cell #2 doesn’t have, and if that mutation causes E coli cell #1 to have lower reproductive success than E coli cell #2, then E coli cell #1 will be selected against and probably go extinct, depending on how big the fitness loss is.
At no point do we need to assume anything is identical for the lifestyles between E coli and Homo sapiens.
Well, on another note I submitted the claim that 300 amino acids out of roughly 500 in the F-type ATP synthase subunit beta, have been conserved across the diversity of life, to a test.
I sampled broadly across the diversity of life known to have F-type ATP synthases, and selected approximately one hundred F-type ATP synthase subunit beta protein sequences for alignment (there are tens of thousands on Uniprot).
I chose sequences from dusins of species of bacteria from several large orders, from plants, animals, sponges, fungi, choanoflagellates and other odd and obscure eukaryotes.
A sequence alignment of all these F-type ATP synthases subunit beta proteins shows that even for the roughly one hundred sequences I chose, only ~5% of amino acids are conserved. It’s probably even less than this if all protein sequences from F-type ATP synthase beta subunits from Uniprot were submitted to an alignment. I excluded all fragments, or sequences that weren’t annotateted as “complete”, and only chose those annotated as F-type ATP synthase subunit beta.
Link to alignment: http://www.uniprot.org/align/A2018061983C3DD8CE55183C76102DC5D3A26728B1F7B46K.
Not that any of this really matters all that much to the debate here. Whether 300 or 36 amino acids are conserved across the diversity of life has little bearing on how broad the base of the hill is (though it does matter to how broad the peak is), how many such hills there are, how tall it is, whether it overlaps hills that exhibit other functions, etc. etc.
Generally the rule is that it doesn’t matter how conserved it is, there probably is some obscure species out there making a living with the amino acid you think is essential, mutated to something else.
Life, uh, finds a way.
Huh? Why?
But you’ve not done that. An explanation that can be applied to anything is not an explanation. It’s a cop out.
Here the double standard is revealed once again. To Bill Cole it is entirely acceptable that “Design” can operate exclusively by “inference to the best explanation”, but evolution must provide “demonstration”.
It’s 300 preserved out of a larger number of AAs, Bill, which means that the rest change. Thus, we’re observing the sequences dancing around a local minimum (I never said that the AAs were dancing, I clearly referred to the sequences).
Whether it may or may not be a challenge to explain how the ancestral sequence got into that local minimum doesn’t change the fact that you’re mistaking a dance around a particular local minimum for the paths towards any-and-all local minima. I’d prefer that you understood this, so that you stopped misusing the data. Do you finally understand the problem? If not, why not? What’s unclear?
Entropy,
You’re observing a sequence. Here we have common ground. How do we determine that observed sequence is dancing around a local minimum?
I thought you said that we’re observing two sequences that share 400 AAs out of some larger number of AAs. Didn’t you? If so, then I told you already. Since the sequences share a number of AAs beyond what would be expected by chance, they’re homologs. Meaning that they diverged from a common ancestral sequence. Since they tend to preserve a high proportion of AAs, that indicates that they’re somewhat stuck in a local minimum, which also means that they might be performing either a similar, or the same function. The changes in AAs we see represent changes that don’t take the sequences far from the local minimum.
You keep repeating that the sequences don’t allow for a lot of changes. Well, that means that the starting sequence, the ancestral sequence, already had the function. That it was already at a local minimum.
Clearer now?
Entropy,
So far so good:-)
Rumraket,
Nonsense.
This is Corneel’s claim.
Rumraket,
Do you assume that a mutation that creates reproductive success or reproductive failure in bacteria will also create reproductive success or failure in humans?
Rumraket,
I did 6 living organisms.
E coli
Human
Mouse
Rat
Chicken
A flower (mouse-ear cress)
Results were worst case 60%+ alignment and best case 98% alignment. The interesting result was between human and e coli 60%+ and humans and flower 60%+ and e coli and flower 60%+.
So if the AAs were identical over 1 billion years you have no information about the base of the hill?
So if the AAs were as different as expected by a random search you have no information about the base of the hill?
Of course the number of substitutable AAs correlate with the base of the hill.
You are missing a major point here: Butterfly eyespots are a novel feature in Lepidoptera. I haven’t bothered to calculate the FI of the developmental machinery that is responsible for it, but we can be fairly certain that it exceeds the 500-bit limit by gpuccio-math: Hence we must conclude it was designed. Yet NONE of the proteins that are discussed in the paper are novel genes; they ALL have previously known functions in insect development. For example, hedgehog is a segment polarity gene, just like Wingless.
Regardless of whether natural processes are responsible for the co-option of an existing pathway or whether the good Designer has decided to recycle some of his oldies, this fact in itself is devastating to the 500-bit limit rule: It negates the requirement that proteins have to evolve de novo to create a new phenotype, and therefore provides an inroad for evolution to inject more than 500 bits of FI at once into a novel biological function.
This is a valid argument, and I don’t think you can ignore it like you do.
Again, why? If there were only single targets, could they evolve then?
Neat! I was delighted to find the mitochondrial sequences* nested in alpha-proteobacteria and the chloroplast sequences clustering separately (just check the “highlight taxonomy” box to visualize them).
Do you know why there is that odd outlier in the tree? It is annotated as an alpha chain (A0A2C6KB02_9BACT). Perhaps a mistake?
ETA * just checked and the gene is nuclear. Interesting, I wonder what the story is with the chloroplast versions.
Not “a” flower. That is a famous model organism that is better known by its official name: Arabidopsis thaliana.
Hmm could be one I added by mistake, I filtered out a lot of sequences when making that alignment (had collected over 300).
You are right, that’s definitely an alpha subunit, I will edit and resubmit. Thanks for catching that. Makes sense that it brings the identity score down to 5% as that’s roughly what is expected by chance for protein sequences that aren’t homologous.
Done. New alignment: http://www.uniprot.org/align/A201806209011D4373CD8E192AC56F039114E7259139B6FH.
Indentity score climbed to ~14.3% with the alpha subunit sequence removed from alignment.
There is probably a correlation there, but we don’t have to assume it will always be the case, no.
I’m sorry but what I wrote perfectly encapsulates how you argue in this thread.
Regardless of whether they are conserved or not we don’t get information about the base of the hill. They could be at a local optimum on a hill with a very broad peak. The fact that this is one possibility out of many means we can’t claim to know how the base of the hill looks without doing actual experiments on the protein.
How so? What is the relationship between the base of the hill and the number of substitutable AAs? Explain this.
Rumraket,
If the 50 AA protein can tolerate only a few AA substitutions and still maintain function then we have a very small base.
If 50 AA can tolerate lots of AA acid substitutions then the base is much wider.
The hill simply represents the quantity of functional sequences that are similar.
Rumraket,
Has it occurred to you that it is because I am dealing with lots of unsupported claims? If you can show evidence of an exception to simply pushing against unsupported claims I will accept your criticism.
Corneel,
Not at all. The 500 bit rule starts with observed origin of the genetic information. In the case of WNT and Shh these would be somewhere around early multicellular life.
Although the WNT and Shh ligand perform a similar function of binding to a receptor protein and initiating transcription that does not mean they contain the same FI and in fact they must contain different FI or they could not work as initiators of separate signaling pathways.
OMagain,
I honestly think the whole concept of functional sequences evolving is problematic. The problem simply gets worse as you add FI.
Corneel,
Thanks 🙂
That is correct.
Since there is a 1:1 mapping from sequence to altitude, therefore (mathematically) the base of the hill must be at least as wide the summit. Thus we can say that the base is as wide or wider than the degree of constraint that we see at the summit…
…but how much wider, we cannot say. So your “correlation” will be poor or non-existent. All we know is that width-at-Base > width-at-Summit.
Gpuccio is insisting that the terrain is completely flat, and that the hills have vertical sides; he denies the existence of slopes (unless someone explicitly surveys a slope for him, in which case he has promised to move the goalposts…). He also ignores the fact that there are other peaks, some of them tall enough, others even taller, than the activity he is observing.
So yeah, in gpuccio-land, the area at the summit is equal to the width at the base. An awesome correlation. The data, however, show this to be a convenient (for ID) fantasy.
DNA_Jock,
Why would it not tell us that the hill is small therefore the base is small?
DNA_Jock,
From reading his critique of the TSS I think the above is due to you’re misunderstanding but please support where he made the above claims if I am wrong.
I would also ignore because your claim that this is a fact is false.
Oh look. “Cell cycle control both in embryo development and mature animals” has suddenly vanished from the description of Wnt function. I wonder why that is.
So how much FI was added to this new function from “somewhere around early multicellular life” to modern humans? I’d say exactly zero, because this function didn’t change. But … what FI did the Designer introduce in those information jumps then? And have you thought about how many proteins can potentially act as a ligand to a receptor protein?
There is not one function, so no single measure of FI. For the function “binding a receptor” the FI will be more or less equal. For the latter functions “binding the frizzled receptor” and “binding the patched receptor” the FI is different, because the receptors co-evolve with their ligands. But to use those measures as an argument against the plausibility of evolution of these pathways presupposes the existence of those specific receptors after the fact.
Ack! It is telling us that the hill has a sharp peak. From the sharpness of the peak, you and gpuccio are trying to draw a conclusion about the area of the island that this peak sits on. It is deranged.
How does one “add FI” and to what is it added?
Please think about that and give your best answer.
How does one “inject FI” and into what is it injected?
Please think about that and give your best answer.
😉
Then how do you explain the following?
There has to be a reason to think that “shared common ancestry” is the best explanation. What is that reason?
Would I say it if it were not true?
I certainly believe it is true. And Joe isn’t claiming that evolution can create FI.
This is nonsense. You don’t know what the FI is and you haven’t calculated it. You haven’t defined either the function or the threshold. The FI of the population? What on earth is that? What is the function of the population and what is the threshold.
Natural selection leads to MORE sequences that perform the function not fewer and thus the FI would actually DECREASE not increase.
Nope. What you wrote is nonsense.
Firstly, one does not need to be able to calculate the FI of an element in order to be able to describe situations in which it has increased, or situations in which it has decreased. We’ve been over this before – we were even talking about proportions…
I took Joe to be referring to the average FI of the population, calculated (as we do) using the observed activity as the threshold in every instance.
The FI calculation does not depend on how many sequences in existence exceed the threshold, but on how many se
ETA” sequences in the sequence space do.
posted in error… sorry
That’s not the way you calculate this.
You start with the set of all sequences, and consider drawing sequences from that set randomly. Set the threshold of “function” as some value,
. If the fraction of all sequences that have function at or above that level is 
, then the value of FI (per Szostak and Hagen) is 
.
Szostak and Hagen consider a sequence from the population after adaptation has been favored (by natural selection). If all of these are above the threshold, then the average FI is
. If a fraction 
, which is greater than 
are above the threshold, then the increase in FI would be 
which is positive. If 
then the increase in FI has been zero.
DNA_Jock,
How do you determine whether your on a peak or an island?
Corneel,
There is a difference between the WNT signaling pathway and the WNT ligand. BIG DIFFERENCE. The same goes for the Shh ligand and the Shh signaling pathway. These pathways are interdependent and are a big part of embryo development.
Why do you believe this?
Is there an explanation how randomly changing DNA of one gene changes the other or is this an evolutionary inference 🙂
So you start with coevolution and now here is the simplified WNT Sonic pathways.https://goo.gl/images/X1LzS3
Joe Felsenstein,
Circular reasoning?
It’s your method. You tell me.
I don’t think that you can tell anything about the lower slopes if you only look at the peak. You and gpuccio claim otherwise. Educate me.
Here’s a toy example, so we can work through the numbers together:
Being a thorough chap, I sequence a billion extant variants of a particular sequence. I find 800,000 unique sequences. How big is the zone of attraction for this sequence?
Show your working.
DNA_Jock,
The peak and island discussion is not coherent for interactive proteins like the F1 beta chain of ATP synthase. The key measurement we can make is AA conservation to determine FI. A measurement above 500 bits infers design.
How do you know that? How do you know you’re not just looking at the peak? Why do you think the sequence variants we see extend down to the base of the hill?
Look at this picture.