|
Author
|
Topic: Seemingly Unevolvabe Biochemical Systems
|
Sage Ross
Member
Member # 1266
|
posted 21. April 2004 13:34
Hi all,
I’m a student at U. of Oklahoma, and a member of our IDEA (Intelligent Design and Evolution Awareness) Club. Next week I will be giving a presentation on biochemistry and intelligent design. My basic premise is that the question of design in biochemical systems is not obvious, and that instead of simply dismissing speculations or conjectures of design as ignorant or already invalidated by the encompassing truth of evolution, they are worth investigating.
The meat of this presentation is presenting biochemical systems that at first glance seem to be irreducibly complex and unevolvable. I will try to explain the definition of irreducible complexity that Behe developed in his responses to criticism the blood clotting system as IC, which is to say, the more neutral (or unfavorable) mutations that would be necessary to evolve a protein or system of proteins from a functionally unrelated one, the larger the probabilistic barrier to evolution. My argument will then be that these systems are interesting research problems to be investigated to provide evidence one way or the other about their evolutionary or designed origin.
However, as an amateur in biochemistry (it’s my major, but an undergrad understanding of biochem doesn’t get you very far), I don’t know what has already been said or done about these systems. They seem to me worth investigating in the context of testing intelligent design, so I thought I would summarize the systems I will be presenting, and why they seem IC or unevolvable. Any and all feedback or criticism will be greatly appreciated. If you notice why any of them are not IC, or where I get facts wrong, please tell me.
System 1: GroES/GroEL type proteins (Hsp60 and Hsp 10 heat shock proteins)
This pair of proteins is universally present in E. coli and other bacteria, and related proteins are present in many organelles and perhaps all eukaryotes, and the pair are essentially a machine for folding unfolded proteins. One thing they do is allow bacteria to survive heat-shock: when proteins are damage and partially denatured by heat, they utilize ATP energy to fold them back. As I understand, the bacteria that have them do not survive without them, as they not only refold damaged proteins, but originally fold proteins that are necessary for survival. Many proteins simply do not fold to their correct useful state without these or other chaperone proteins to help them fold. Not only that, but if elegance, beauty and coolness are at least implicit indications of design, then the GroE proteins have evidence for ID in spades.
System 2: Shape-shifting protein channel
The protein channel of the type that transports newly synthesized (partially folded) proteins from the ribosomes on which they are put together into the ER is universal. In order to selectively transport proteins out of the cytosol (without simply being permeable to all small proteins and other molecules, the proteins change shape; the pores grow large enough to accommodate alpha helices. This change happens in response to the signal sequence at the beginning of the proteins bound for export. It seems that in order for signal sequences to be useful (they are cleaved after transport before the proteins achieves their mature, useful states), a mechanism of transport and cleavage must exist. But in order for a selective, shape-changing protein channel to be advantageous, it must be able to distinguish which proteins need to be transported and which don’t. Can a series of selectable mutations create this system? What selective advantage does the shape-shifting protein channel have unless there are proteins with signal sequences (whose function ultimately depends on not having those sequences)? If we don’t already know, is it worth investigating?
System 3: The tail lysozyme of bacteriophage T4
Like the flagellum, the drill-like protein used by this virus to puncture cell walls has a striking resemblance to human inventions that serve analogous functions. In this case, the structure of the protein is so unlike any other proteins that I know of, and so specific to its function, that it seems difficult to imagine an evolutionary scenario where the function evolved gradually from another function. I haven’t seen any other proteins with structures that seem like they could be co-opted to form this sort of puncturing device that could evolve into the seemingly designed one that exists today. I really don’t know much about how this works, so I’m just throwing it out there to see if anyone has any interesting comments.
-Sage Ross
IP: Logged
|
|
Scott
Member
Member # 1222
|
posted 22. April 2004 01:15
fyi, afaik Behe does not claim that the clotting system is IC.
cheers
IP: Logged
|
|
Sage Ross
Member
Member # 1266
|
posted 22. April 2004 01:38
He may no longer hold that the blood clotting cascade is IC, but he does formulate what seems to me the most useful and relevent definition of irreducible complexity in "In Defense of the Irreducibility of the Blood Clotting Cascade: Response to Russell Doolittle, Ken Miller and Keith Robison" from July 31, 2000.
At the least, the definition of IC he develops in response to Keith Robison seems like a pretty straightforward way to evaluate proposed evolutionary pathways, even if it doesn't conclusively prove the design or unevolvability of a protein system (since there might be some unimagined pathway that involves only selectable mutations). Also, it accounts for the common criticism that IC structures can evolve; a system that is only 1 degree IC could certainly evolve, and with more degrees it is still possible, but increasingly unlikely. Has anyone actually done any work figuring out the degrees of IC represented by evolutionary senarios and compared them with phylogenies? I'm sure this stuff has been hashed out on this site before, so you'll excuse me if I don't provide novel insights. Any good past discussions you can point me to?
[ 25. April 2004, 17:47: Message edited by: Sage Ross ]
IP: Logged
|
|
rafe gutman
Member
Member # 134
|
posted 29. April 2004 18:03
hi sage,
it sounds like you think that IC systems with only a few parts could have evolved, but not systems with a large number of parts. is the only difference between evolvable IC systems and unevolvable IC systems quantitative, or is there a qualitative difference?
if you're looking here for a lot of material on irreducible complexity, you'll probably be disappointed. this is dembski's most recent musing on IC. i've noticed that ID proponents don't really want to discuss IC anymore, but you'll find critics aplenty who do.
IP: Logged
|
|
Sage Ross
Member
Member # 1266
|
posted 30. April 2004 19:34
rafe,
My thoughts on IC and evolvabitility may not be up-to-date, but yes, essentially the difference is quantitative for what I'm talking about. But the important quantity is not the number of parts, it is the number and liklihood of the non-selective mutations necessary to produce the IC system (if any).
Of course, just because a system seems irreducibly complex and unevolvable by a series of beneficial mutations doesn't mean there isn't some unthought-of pathway. And just because there are a small number of non-selective mutations required for a given IC system to arise, doesn't mean it didn't evolve. But if a very unlikely set of mutations would have been necessary to produce an IC structure, it becomes a piece of evidence suggesting design (depending on just how unlikely the evolutionary pathway is compared to the timeframe in which that system arose).
Thus, IC itself is not evidence that a system could not evolve, but a starting point that suggests the system's origin is worth investigating from a design vs. evolution perspective.
I'm mainly looking for some feedback and thoughts. Thank you for the Dembski post... I haven't read anything as recent by Behe or Dembski, so hopefully it will help me clarify some of my thoughts on the subject.
IP: Logged
|
|
Art
Member
Member # 179
|
posted 01. May 2004 15:30
Hi Sage,
A few random comments about some of the "items" you mention:
1. The GroEL/GroES system is but one of a myriad of chaperonins. Some have claimed that they may be IC, but they are better examples of the broad utility of cooption. PubMed and BLAST searches reveal that the motifs and molecular themes of different chaperonin systems occur in many, many places (I recall one paper that linked either HSP60 or HSP70, I forget which, to the ancestry of F1 ATPases). But beyond this, and much more tangibly, we see in the mryiad of interactions involving chaperonins a decidedly low-information (zero CSI) network that can easily (and visibly) evolve so that new functionality can be added to basal IC systems.
2. The SRP-associated translocon you refer to is not the only protein-transporting translocon "out there". Much can be learned (interestingly, stuff that relates tangentially to viral "injection" systems) by comparing and contrasting different translocons.
Also on this subject, you did not mention the crucial role of Bips in ER-associated protein transport. Bips are essentially HSP70's, and stand as an interesting example of how the parts of IC systems can be coopted for rather different functions (of course, this grants that the "normal" function is that first mentioned in these discussions - but that's a nit-pick).
(Someday, it would be fun to explore the idea that chaperonins are one of the, if not the, primordial polypeptide-associated functionalities.)
[edited to fix one of what are probably several typos]
[ 01. May 2004, 19:12: Message edited by: Art ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 01. May 2004 23:24
Scott wrote quote:
fyi, afaik Behe does not claim that the clotting system is IC.
I'm afraid that's a bit of revisionist history. In DBB Behe originally argued that the vertebrate blood clotting cascade is irreducibly complex. In 2000 he attempted to defend that position against criticisms. Finally, in 2002 under questioning from Ken Miller, Behe apparently retreated a bit from that position, though it's hard to tell from context just how far he retreated. He said that the blood clotting cascade is not irreducibly complex "In the same sense that a rattrap is not, that's correct." Since I'll be darned if I know what "in the same sense that a rattrap is not" actually means since rattraps did not occur previously in that conversation (though mousetraps did, of course), I can't tell what Behe now believes about the vertebrate blood clotting cascade's irreducible complexity. Is there a difference between the ICness of mousetraps as opposed to rattraps? I dunno.
RBH
IP: Logged
|
|
Scott
Member
Member # 1222
|
posted 02. May 2004 12:37
I was not speaking historically, so it is hardly revisionist history. Please delete or modify your post accordingly. Thank you.
IP: Logged
|
|
Argon
Member
Member # 276
|
posted 02. May 2004 13:37
Sage Ross writes:
"At the least, the definition of IC he develops in response to Keith Robison seems like a pretty straightforward way to evaluate proposed evolutionary pathways,.."
IC was initially defined as a means of identifying intelligently designed systems without having to resort to a complete, highly-detailed analysis of all possible evolutionary pathways (which, for most cases, is not technologically possible). With this original definition, one simply evaluates a system in its present context and classifies a system as "IC" on the interactions between a system's "parts".
Under the original definition, much of the blood clotting cascade is certainly "IC". The test? This was clearly specified in Behe's DBB: Simply remove or significantly alter one of its parts and see if fails to properly operate as a whole.
In contrast, IC-mark II requires the evaluation of possible evolutionary pathways before any system can be classified as "IC". By rights, this version shouldn't be called "IC" because it is a separate beast altogether and because there is no developed link with originally formulated, IC version-I. Version-I is all about the current status of a system: Mark-II depends on determining the evolvability of a system. Thus the blood clotting cascade and the flagellum can't be classified as "IC" under the second definition because nobody has been able to evaluate the possible evolutionary pathways because of technological limitations. IC-mark II is anything but straighforward to implement.
So when Scott suggests that Behe doesn't think the blood clotting system is IC, that is because Behe is mixing non-interchangeable definitions (see here).
[ 02. May 2004, 13:43: Message edited by: Argon ]
IP: Logged
|
|
Sage Ross
Member
Member # 1266
|
posted 02. May 2004 14:36
Thanks for your comments, especially Art.
I try to use the second definition of IC (as it seems more useful in getting at the heart of the debates), and so I call systems "seemingly" or "apparently" IC. The problem is, indeed, that IC-II is not nearly as easy to work with as IC-I, but it still seems like a workable concept. If you think it is not, why not?
Argon, what are the technological limitations that prevent the analysis of proposed evolutionary pathways for the blood clotting cascade?
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 02. May 2004 15:44
Scott wrote quote:
I was not speaking historically, so it is hardly revisionist history. Please delete or modify your post accordingly. Thank you.
Sure, if someone can tell me what the critical difference between 'IC as in a mousetrap' but 'not IC as in a rattrap' is. The rattraps we used to use on the farm were of exactly the same design as the mousetraps, just larger, and to my knowledge Behe still uses a mousetrap as an example of ICness. Hence his remark that the vertebrate blood clotting cascade is not IC "In the same sense that a rattrap is not, that's correct" seems to me to suggest that he still deems it to be IC.
Alternatively, point out where Behe unequivocally abandoned his claim about the irreducible complexity of the vertebrate blood clotting cascade, and which version of IC he abandoned. I may well have missed it.
RBH
IP: Logged
|
|
Argon
Member
Member # 276
|
posted 02. May 2004 22:55
Sage Ross writes:
"Argon, what are the technological limitations that prevent the analysis of proposed evolutionary pathways for the blood clotting cascade?"
That would depend on the level of detail required. The basic problems come from insufficient data and under-specified models.
The mechanisms and interactions between the components of the blood clotting system are not sufficiently characterized to model clotting as it exists even today in very much detail. What we have are a few operational parameters, a bunch of proteins and a subset of what we think are key interactions among some of the components of the system. But there are many parts missing. Thus, if we wanted to model the system in such a way as to assess 'fitness' or even some functional parameters at the level of amino acid variation or changes in the timing or expression of components, we could not do it --- And that does not even include interactions with other blood components or physiology that could also occur over the course of evolution. In short, it would be a GIGO exercise.
The best that can be done at this point is examine the system as it exists today in a great number of related species and try to use inferred phylogenies to reconstruct approximately when the components might have arisen. Unfortunately, that provides us with little detailed information about how each component was 'conscripted' into the clotting pathway. One might be able to say (maybe) that component-X was added in lineage-Y sometime after it split from lineage-Z, xx-million years ago, but not know a whole lot more detail about the system's operational parameters at that time (such information was probably obliterated over time). While we can sometimes determine some very general details about the evolutionary history of a biochemical system, we would have a very hard time developing or evaluating scenarios like: "Component-X's gene was frameshifted in 339,003,333 BC which increased its affinity for component-Y by 100-fold and consequently allowed clots to withstand 10 kpascals of static pressure in an organism with a blood pH of 6.8 in its capillaries but pH 7.0 in its central arteries."
Actually, I agree that IC-mark-II provides a better set of criteria for getting at the heart of the debate. First, it is about 'evolvability'. Second, it illustrates how very difficult it is to definitively evaluate the question. For example, we do not know if any natural, biological system fulfills all the criteria. The down side is that it is not a terribly novel approach or concept because what it boils down to trying to evaluate the probability of any particular evolutionary step... something that has been argued about for years.
(Aside: Gads, now I've something else to bitch about: IC-mark-II not only leads to confusion with terminology, it is also a recycled argument with a new name.)
[ 02. May 2004, 23:12: Message edited by: Argon ]
IP: Logged
|
|
Scott
Member
Member # 1222
|
posted 03. May 2004 04:36
quote:
So when Scott suggests that Behe doesn't think the blood clotting system is IC, that is because Behe is mixing non-interchangeable definitions.
It's not a suggestion, it's a statement of fact. You even seem to be agreeing with me, in a sort of back-handed way.
quote:
Under the original definition, much of the blood clotting cascade is certainly "IC". The test? This was clearly specified in Behe's DBB: Simply remove or significantly alter one of its parts and see if fails to properly operate as a whole.
"Much of..." does not make an IC system, and "fails to properly operate as a whole" is not any part of the definition of an IC system.
IP: Logged
|
|
rafe gutman
Member
Member # 134
|
posted 03. May 2004 18:08
scott, what do you think was the purpose of this article by behe, which RBH linked to in his post?
In Defense of the Irreducibility of the Blood Clotting Cascade:
Response to Russell Doolittle, Ken Miller and Keith Robison
here's the first sentence:
quote:
In Darwin's Black Box: The Biochemical Challenge to Evolution I devoted a chapter to the mechanism of blood clotting, arguing that it is irreducibly complex and therefore a big problem for Darwinian evolution
sage, i don't think you should use behe's 2nd definition of IC (IC-II). most people ignore it because it's tautological. it's meaningless to say that an IC-II system is unevolvable if the definition of an IC-II system is one that requires unselectable steps. if you choose to use that as your definition of IC, then you must recognize that no IC-II systems have yet to be discovered. how do we know that the blood-clotting, bacterial flagellum, or other complex systems require unselectable steps in their evolution?
i'm curious about this statement of yours:
quote:
Thus, IC itself is not evidence that a system could not evolve, but a starting point that suggests the system's origin is worth investigating from a design vs. evolution perspective.
so how do you propose to investigate a system's origin from a design perspective? what studies can you perform on any biological system to advance the notion that it was created by an intelligent agent? for example, suppose i gave you 1 million dollars to study ID. what would you do?
IP: Logged
|
|
Scott
Member
Member # 1222
|
posted 03. May 2004 19:27
quote:
scott, what do you think was the purpose of this article by behe, which RBH linked to in his post?
I don't care what the purpose was, it's irrelevant. What do you want from me? Did you not read the links RBH provided?
Here it is in his own words, from his own post:
quote:
Finally, in 2002 under questioning from Ken Miller, Behe apparently retreated a bit from that position, though it's hard to tell from context just how far he retreated.
You can read Behe's own words by following the link RBH provided.
Sage, my apologies. This thread seems to be getting hijacked. You wish to discuss evolvability, or lack thereof, and not IC'ness, is that correct?
[ 03. May 2004, 19:37: Message edited by: Scott ]
IP: Logged
|
|
|
Printer-friendly view of this topic