I apologize if I’ve been unclear, but I thought it would be fairly obvious that if one wants to assess the functional relevance of a particular quantity of genetic material, one should delete that genetic material and measure the amount of function remaining in the resulting mice.
Call me old-fashioned, but I think that actually quantitating function is the best way to quantitate functional information.
The problem we are not just discussing how to measure functional information , but how to measure change in functional information.
You and I might think that, if a sequence of a mouse’s genome was duplicated and then diverged so that one performed a new function that was not present in the original genome, this represents an increase in “functional information.”
But would ID proponents agree? Obviously not, or they wouldn’t still be saying that FI can only be produced by intelligent design.
Point well taken. I just wanted to emphasize that even if we ignore substitutions or small indels, recombination still results in information that simply was not there to begin with.
Merry Christmas, and a happy new year.
Your point is well taken as well.
It’s important to note that neither @Giltil nor @gpuccio have addressed both of these increases at the same time, although they both occur in just weeks, they both increase what they narrowly define as FI, and they both increase function measured directly.
Possibly to assist anyone who believed he’d read them both.
Ah, right. Well, I’m still reading up on B cell development and V(D)J recombination so it’s seeming like a bit of a black box. What you said makes perfect sense, but 1) I wasn’t sure if there was something between the genetic material and function that would interfere 2) I’m not sure what we would delete. Would we just delete individual V, D, or J units?
I am back!
This whole conversation was put forward by @Mercer who claimed that the process of VDJ recombination was able to produce high FI. I disputed this claim, noting that “V(D)J recombination doesn’t produce high FI because most of the genetic information necessary for implementing the function preexist in the stem cells that develop into B cell”. The thing to see here is that the claim that most of the genetic information necessary for implementing the function preexist in the stem cells is not an hypothesis anymore, not at all. It is common knowledge since the discovery by Susumu Tonegawa of the genetic principle for generation of antibody diversity that earned him the Noble Prize in 1987. The figure below may help the uninitiated to grasp this point.
Now, I would like to draw attention to another point that also shows that the process of VDJ recombination doesn’t generate high FI. Let’s take a common lymphoid progenitor (an early pro-B cell) and ask ourselves what is the probability that, through the process of VDJ recombination, it will give rise to a pre-B cell able to express a bona fide immunoglobulin heavy-chain. The answer is: very high, roughly 50%. So what does this result tell us about the process of VDJ recombination? It tells us that it is a highly constrained process (a process in which, given state A, state B is highly likely), and, as such, a process unable in itself to create high FI.
Now some readers here may find the situation quite bizarre. Indeed, I plainly recognize that the FI of say an immunoglobulin heavy-chain is high while at the same time I contend that the process that gives rise to it doesn’t by itself generate new, high FI. Why is this? To see what’s going on here, think, for example, of a fully automated car assembly line. The cars at the output have high FI. But of course all this FI is not created de novo from nothing. The whole system is programmed, by design, to produce cars. IOW, it is a highly constrained system that can’t help but produce cars and where the information require for this result is already present at the input.
To @Mercer: I’ve already fully recognized that the process of somatic hypermutation/antibody maturation is able to generate low but non trivial level of FI, say more or less 50 bits of FI. And you know this perfectly well. So, please, stop blaming me for evading the issue of somatic mutation. It’s really absurd of you, not to say malicious, especially considering that it was you who at the beginning blamed me for having only considered the process of somatic mutation and not the process of VDJ recombination! To warm your memory, that’s what you told me some days ago: “Purely natural processes are actually capable of producing high FI in that antibodies are evolved in only two weeks. Both you and gpuccio have tried to pretend that only one process (somatic hypermutation) is producing FI, while ignoring V(D)J recombination.”
So neither new, nor high FI? Do you really mean to say there is no new information in the final immunoglobulin molecule?
On a related note, FI is calculated with respect to some minimal threshold for function. Only when you know the number of sequences that meet the minimal threshold, in relation to the total number of possible sequences of equal length, can you calculate the FI. That means we first have to decide on what counts as the minimal threshold for function. It seems to me it would only make sense to set it at a level of immunoglobulin binding activity that is high enough to constitute a successful immune response(aka, immunity).
I don’t believe you’ve done the work of actually determining the number of possible sequences prior to VDJ recombination, that would meet the threshold.
No, it doesn’t tell us that because you can’t predict with a 50% reliability what the exact sequence of the final immunoglobulin molecule will look like. The mere fact that it will be presenting one is not the point of contention, it’s the unique and unlikely sequence of it, the result of recombination (and somatic hypermutation) that is.
You seem to be saying that you can “predict” the outcome of card draw like this: A card will be drawn with high probability!
If we assume that he was using the same definition of functional “information” as you are (which is far from a given), then what this demonstrates is that the evolution of new functional proteins does not require the creation of new functional information, which completely refutes the entire premise of ID Creationism.
So, well done.
Welcome back. Did Santa Claus by any chance bring you that definition of “new information” and/or an example in a genome which would count as “new information” for you?
Let’s try another approach:
How do you quantify the amount of FI in a genome or protein? If you can’t measure the amount how do you know if the FI increased, decreased, or stayed the same with each new generation?
For a protein, by using @gpuccio’s methodology, that is by assessing conservation through deep time.
I respectfully disagree. I think @gpuccio’s methodology is a very clever way to assess the amount of FI within some proteins and I haven’t seen good rebuttal of it until now at this site.
Then read the link in my post.
That is utterly false, Gil.
My point is that all of those mechanisms, not just VDJ recombination, produce high FI. You are unwilling to make the calculation for all the mechanisms and you are misrepresenting my position.
And I, in turn, have proposed looking at this experimentally, an approach you clearly reject. Why is that? What are you afraid of, Gil?
No, Tonegawa’s work says nothing about “most” of the genetic information.
Tonegawa was quite clear:
“In addition, mutations are somatically introduced at a high rate into the amino-terminal region. Both somatic recombination and mutation contribute greatly to an increase in the diversity of antibody synthesized by a single organism.” --Susumu Tonegawa
Also note that you are moving the goalposts from functional information, the subject you claim to be interested in, to genetic information. Why are you doing that, Gil?
Why are you trying to distract from considering an experiment to see how necessary all of this information really is?
Why are you not doing any math?
It’s not about concessions, Gil, it’s about your fear of actually knowing through experiments whether the FI is already there and how much of it there is. So conceding that it is “low but non trivial” isn’t grappling with the actual issue.
Gil, what does your hypothesis predict would happen to function if we delete most of that FI you hypothesize already exists?
I’m not sure that I can come up with any hints that are more blatant than that.
Your fear of looking deeply is palpable from across the ocean.
Then apply it to all of the mechanisms that create antibody function.
You don’t seem to have any faith in it.
I don’t recall that @gpuccio ever restricted the application to deep time, and as someone who has spent decades studying how changes in sequence alter protein function (measuring function directly), I see no reason why it would not be restricted.
Unless you’re trying to avoid applying it here…
Let me make this more simple, Gil. What you’ve “recognized” is a testable hypothesis. Let’s test it together, using genetic engineering of stem cells.
Why aren’t you interested in that? Where is your curiosity?
Okay. Let’s try to see what is the amount of information (I) produced by a stem cell that differentiates into a naive B cell producing an antibody with a given specificity S.
We have (I)=-log2( P ), with P being the probability that the stem cell gives rise to a naive B cell producing an antibody with specificity S.
Given that estimates of the theoretical diversity of the human naive B cell repertoire range from 10^12 to 10^18, it can be said that P range from 10^-12 to 10^-18 and, as a result, that (I) range from 40 to 60 bits. Note that this range for (I) is an overestimation, for I assume here that there exists only a single antibody with specificity S in the whole repertoire, which in most cases is obviously false. Note also that (I) should not be confused with FI, for FI is related to the improbability to generate the antibody randomly from an unrelated state, which is not at all the case with the process of VDJ recombination.
Bottom line: the maximum amount of information that the process of VDJ recombination is able to produce range from 40 to 60 bits, far, far away from the 500 bits required to draw a design inference. IOW, neither this process nor the process of somatic hypermutation invalidate ID.
