The Signature in the Cell Challenge
by BradfordBioLogos’ Falk on Disco’s Meyer is a Panda's Thumb entry which refers to Darrel Falk's review of Signature in the Cell by Stephen Meyer. There are plenty of unsupported assertions in the comment section. Falk is a biologist and is not persuaded by Meyers. Arguments from authority… there's a trendy term for that. Hmmmm. That's it! Boring.
Let's go where reviewers are reluctant to trod. Here's a challenge for those of you who agree with Falk. Cite something Meyer said in his book, show that there is a problem with Meyer's citation of facts or his reasoning from them and then back up your claim with links to papers supporting your position. This might be difficult for many of you because it requires that you actually take the effort to read Signature in the Cell. Then you would have to use your brains to analyze a specific passage from it and support your arguments with outside references. But why not break from the mold and give it a shot. Don't be afraid to lock horns with harmless tards. You have nothing to lose but your pride.
December 31st, 2009 at 6:27 pm
[Nick's first post, which I, Bilbo, sent to the memory hole by mistake]:
Submitted on 2009/12/31 at 6:12pm
See, this is good old-fashioned TT fare here.
I have read Signature in the Cell. There are many problems with the argument, here is one. Meyer says that "information" — sequence-specific function — is densely concentrated in the DNA genome:
"Thus, far from being dispersed sparsely, haphazardly, and inefficiently within a sea of nonfunctional sequences (one that supposedly accumulated by mutation), functional genetic information is densely concentrated on the DNA molecule." (p. 461)
"Far from containing a preponderance of "junk" — nonprotein-coding regions that supposedly perform no function — the genome is ted by sequences rich in functional information." (p. 461)
Furthermore, says Meyer, not only is this established truth, but it is a prediction of ID theory, and furthermore it was predicted by ID advocates a decade or more ago:
"The genome does display evidence of past viral insertions, deletions, transpositions, and the like, much as digital software copied again and again acumulates errors. Nevertheless, the vast majority of base sequences in the genome, and even the many sequences that do not code for proteins, serve essential biological functions. Genetic signal dwarfs noise, just as design advocates would expect and just as they predicted in the early 1990s." (p. 461)
However, at numerous places in the book, Meyer notes (correctly) that repetitive sequences have little information:
"Since information and improbability are inversely related, high-probability repeating sequences like ABCABCABCABCABCABC have very little information (either carrying capacity or content). And this makes sense too. Once you have seen the first triad of ABCs, the rest are "redundant"; they convey nothing new. They aren't informative. Such sequences aren't complex either. Why? A short algorithm or set of commands could easily generate a long sequence of repeating ABCs, making the sequence compressible." (p. 107)
Unfortunately for Meyer, he seems to not realize that 40-50% of the human genome (and most animal genomes of similar size) consists of LINEs, SINEs, segmental duplications, and other repeating elements. As documented here:
http://www.ncbi.nlm.nih.gov/bo...
In other words, there is no way that in "the vast majority" of the genome genetic information is "densely concentrated" — as proven by his own arguments!
QED.
[And this is Nick's second post]:
Also posted relevant graphic to PT: http://pandasthumb.org/archive...
Comment by nickmatzke — December 31, 2009 @ 6:27 pm
December 31st, 2009 at 7:37 pm
Wow I was unaware of the DNA sequence ABCABCABCABCABCABC.
I believe what Meyer is saying, nick, is that the parts of the genome that actually do something are not haphazardly scattered- for example HOX gene clusters.
Comment by ID guy — December 31, 2009 @ 7:37 pm
December 31st, 2009 at 7:39 pm
So 50-60% is merely a majority and not a "vast" majority and therefore Meyer's crime was the use of the word "vast"?
Or was it? I seem to recall recently that a 60% majority wields pretty "vast" power, at least in the US Senate.
"A" for effort Nicky. What else you got?
Comment by chunkdz — December 31, 2009 @ 7:39 pm
December 31st, 2009 at 9:46 pm
Seriously? So your counterargument is that (maybe) 60% is nonrepetitive, and this wields "vast" power in the U.S. Senate (highly debatable), therefore Meyer is right about the genome?
"Vast" should really be like 90%+ high-density "information". This is clearly what he and the other ID guys actually think — they think there is very little junk DNA and the whole concept is a massive error. The real numbers look like coding + noncoding regulatory DNA + other tidbits = maybe 10% sequence-specific function. The remaining 40% of the genome (the half that is nonrepetitive) is mostly thought to be transposon bits etc. that have decayed so much they just can't be recognized, but we don't have to argue about that bit to show that Meyer' claim that the "vast majority" of the genome has "densely concentrated" information simply can't be right.
Comment by nickmatzke — December 31, 2009 @ 9:46 pm
December 31st, 2009 at 10:37 pm
Nick Matzke wrote:
Before responding to your argument let me point out for the benefit of some that Nick's comments are consistent with good commenting practices and reflect favorably on the author while benefiting the website.
Meyer would have been better advised to omit the word vast. We can quibble about how much is functional and how much junk but whatever the consensus percentage is will not detract from the essential themes of the book. Genetic codes are not the expected outcomes of random chemical reactions and we have no reason to suspect that robust self-replicators are possible without a code enabling the storage and expression of information necessary for function.
If Meyer writes a follow-up book I hope he devotes significant attention to the fact that genomes would have even more junk were it not for error detection and correction mechanisms which minimize errors and safeguard genomic integrity. Error correction is a hallmark of design- one of the more sophisticated genetic properties we have identified.
Retroviral and transposon derived junk indicate genomic predation. They do not mitigate the significance of specification.
Comment by Bradford — December 31, 2009 @ 10:37 pm
January 1st, 2010 at 12:09 am
Two points I want to make:
1) According to the original quote with respect to "vast majority", Meyer says that the "vast majority of base sequences in the genome, and even the many sequences that do not code for proteins, serve essential biological functions." Saying that the vast majority of the genome serves an essential function is not the same as saying that it is "densely concentrated" information. Furthermore, when he says that "functional genetic information is densely concentrated on the DNA molecule", he seems to be saying that there is a much larger percentage of the genome that is functional information (than non-functional), NOT that the information itself is densely concentrated. This makes more sense in the context below…
2) Prior to that "densely concentrated" quote (where the "thus" comes from), we have:
"In Chapter 18, I listed some of the many functions that the non protein-coding regions play. I also noted that — overall — these regions of the genome perform many of the same functions as an operating system in a computer. For example, the noncoding regions of DNA (in concert with other sources of information) direct the expression of bases sequences in the protein-coding regions in DNA. Thus, far from being dispersed sparsely…"
In chapter 18, Meyer references the following paper:
http://www.ncbi.nlm.nih.gov/pu... ("How repeated retroelements format genome function")
My guess would be that Meyer (and from Sternberg's posts on ENV) would not concede that the LINEs and SINEs lack function, and thus simply pointing out that a large percentage of the human (and other) genomes consists of retrotransposons is not sufficient in showing that these are not functional. Yes, the repeated sequences have less information, but they are still functional information, contributing to the high density of functional information within the genome.
However, I do take issue with Meyer saying that "the genome is dominated by sequences rich in functional information." This "rich" adjective gives the impression, as Nick has suggested, that most of the genome consists of highly specified sequences, rather than that much of sequences may be repetitive. Based on the context, it would probably be more accurate for him to have written "the genome is dominated by sequences of functional information".
Of course, this all presumes that the majority of those retrotransposons do in fact have a function, which is certainly in dispute. But it's unfair to suggest that Meyer was completely ignorant of SINEs and LINEs, and that his description of repetitive sequences containing less information somehow disproves his assertion that the "vast majority of base sequences in the genome… serve essential biological functions."
Comment by statemachinist — January 1, 2010 @ 12:09 am
January 1st, 2010 at 12:12 am
NM:
Thank God the NCSE is here to set us deluded, holocaust denying, fundies straight.
Since it is the case that most of the genome is non-information rich, it must be false that the information rich portions of the genome are densely concentrated.
OW! MY HEAD!
Comment by Mung — January 1, 2010 @ 12:12 am
January 1st, 2010 at 12:21 am
Not so fast Bradford.
Anyone, Nick included, can look reasonable if they only include a portion of the facts against which they are asking you to make a judgment.
Nick is talking only about repetitive sequences:
This is known in reasoning as a false dichotomy.
You're being asked to choose between repeating segments and information rich segments. There's at least one more choice. Non-repeating, non-information, segments.
Comment by Mung — January 1, 2010 @ 12:21 am
January 1st, 2010 at 1:08 pm
nick's initial reference seems to be out of date:
It was published before the human genome project published its results.
I, for one, am very unimpressed.
Then nick sez (without any justification):
Only buttheads tell other people what they think nick.
But anyways nick, what do you have to support your position?
Comment by ID guy — January 1, 2010 @ 1:08 pm
January 1st, 2010 at 2:16 pm
My copy of the book is on the way, and I'll read and review it at my blog sometime before the end of the month. One thing I already know is that there's not much point in bickering over the meaning of words like 'vast.' Nevertheless, like Nick, I find the quotes from page 461 to be incredible; eukaryotic genomes are nearly overrun with parasitic sequences, and the enormous variation in genome size among eukaryotes is almost completely accounted for by postulating relative changes in the rate of accumulation of such parasitic elements. (See Ryan Gregory's site and his comments on Nick's post at PT.) If Meyer's claims about genomic structure are typified by those quotes from page 461, then his book is a joke. I'm hoping that it's more interesting than that.
Comment by SteveMatheson — January 1, 2010 @ 2:16 pm
January 1st, 2010 at 2:41 pm
Hi Steve. Meyer imputes design at point of origin. Your point about parasitic sequences is acknowledged but what relevance would this have to tracing the causal pathway to an initial genome?
I just returned from your blog. Some of the blog entries look like good base material for a TT blog.
Comment by Bradford — January 1, 2010 @ 2:41 pm
January 1st, 2010 at 3:44 pm
Bradford, hello and thanks for the welcome. I've previously given TT a good recommendation, and would be happy to participate (on occasion) in discussions here, if that's what is meant by the reference to "base material for a TT blog."
If Meyer "imputes design at point of origin" then his position must be very similar to Mike Behe's. On such questions I am thoroughly agnostic, and though I am a strong critic of the ID movement, I don't have any problem with considering design postulates like that one. On the other hand, Behe's claims regarding genetics and mutation are ill-conceived, and The Edge of Evolution is a shockingly bad book. Meyer will need to do a whole lot better, and the quotes provided by Nick are not encouraging, especially considering that Meyer's book is about genomic evidence. (Right?)
To your question. I would say that pervasive genomic parasitism is of some help in assembling genomic phylogenies, since parasitic elements are thought to be "domesticated" in rare instances. But I certainly agree that explanation of the "causal pathway" that led to the generation of the "initial genome" is a fascinating and daunting challenge, and that SINEs and LINEs are of limited relevance in that task.
Comment by SteveMatheson — January 1, 2010 @ 3:44 pm
January 1st, 2010 at 3:46 pm
Hi Bradford,
Is Meyer only imputing design at point of origin (I assume you mean origin of life), or is he also imputing design later in the history of life?
The major errors (one hopes they are) about junk/parasitic DNA pointed out by Matzke et al. do not bode well for the reliability of Meyer's book. That's what's relevant I suppose.
Comment by IrynaB — January 1, 2010 @ 3:46 pm
January 1st, 2010 at 4:03 pm
Steve:
You are certainly welcome to participate in any thread but what I had in mind was a blog write up by myself on a blog entry (or group of them) from your blog. I would link to one that interested me and which I thought would be of interest to others at TT. I would probably select some quotes of yours and add my commentary. It is something I routinely do anyway but rarely do I have the opportunity to speak with the blogger in advance of my response blog.
Comment by Bradford — January 1, 2010 @ 4:03 pm
January 1st, 2010 at 4:29 pm
Huh. Have you TT guys actually read Meyer's book? There is no way to read it and come away with the idea that Meyer thinks that your various proposed reconciliations would be OK. I.e., if information is dense in only half the genome, that falsifies Meyer's statement that information is dense throughout the genome. As I carefully quoted in my original post:
"Thus, far from being dispersed sparsely, haphazardly, and inefficiently within a sea of nonfunctional sequences (one that supposedly accumulated by mutation), functional genetic information is densely concentrated on the DNA molecule." (p. 461)
"Far from containing a preponderance of "junk" — nonprotein-coding regions that supposedly perform no function — the genome is dominated by sequences rich in functional information." (p. 461)
Not only is this something Meyer said in the book, Meyer very clearly made junk DNA his first, primary, and favorite exhibit in support of the proposition that ID was predictive and scientifically useful.
Why have no ID guys, in all the various pro-Meyer reviews they have put out there, and in all their complaints about the weak critical reaction, never pointed this out as even a potential problem with Meyer's argument?
Comment by nickmatzke — January 1, 2010 @ 4:29 pm
January 1st, 2010 at 5:01 pm
nick,
I just started chapter 14 last night and will get back to the book a little later.
However I did jump ahead to pg 461 and Meyer references chapter 18 to support what he stated.
I have not followed up on his references, but I will when I have a chance.
But as I said before HOX gene clusters are denseley concentrated on the DNA.
And what happens to your "argument" when it is shown that SINEs and LINEs do indeed have a function?
Comment by ID guy — January 1, 2010 @ 5:01 pm
January 1st, 2010 at 5:18 pm
Stephen C. Meyer,
Signature In The Cell
Comment by Pez — January 1, 2010 @ 5:18 pm
January 1st, 2010 at 5:26 pm
Evolution doesn't predict "junk" DNA, evolutionists didn't mean "junk" when they called it "junk", "creationists" misuse the term "junk DNA", there have always been biologists postulated function for some junk DNA, saith the Darwinists.
But lo and behold, they still think junk DNA exists, that junk is still junk and that it argues against ID. Funny world.
Comment by Pez — January 1, 2010 @ 5:26 pm
January 1st, 2010 at 6:17 pm
Well, first, I like Nick Matzke and I share most of his opinions on creationism and ID. But this time, I really hope he's wrong. If "Meyer very clearly made junk DNA his first, primary, and favorite exhibit in support of the proposition that ID was predictive and scientifically useful" then the book is a joke, and it will be remembered as a notable final gasp of the dying ID movement. That would be unfortunate, since I have the impression that Meyer is a smart and reasonable guy.
ID Guy, you should think harder about your Hox gene argument. If Meyer means to say that "functional genetic information is densely concentrated" in "the genome" (human, I'm guessing) then he needs that to be true throughout. Gene-rich islands here and there don't add up to the claims that Nick cites. To argue as you do is akin to asserting that Wyoming is densely populated while pointing at Laramie.
And your claim that Hox genes are clustered is a generalization that is true of a limited number of species. In fact, Hox gene collections are organized in varying ways, ranging from highly organized clusters in vertebrates through almost completely split-up ("atomized") collections in other animals.
Let's hope that Meyer has more understanding of comparative genomics than these comments suggest.
Comment by SteveMatheson — January 1, 2010 @ 6:17 pm
January 1st, 2010 at 6:37 pm
Meyer writes:
I'm not sure this is correct. If I remember correctly, in the human genome, regulatory activity is suspected primarily in the "stable" (i.e, conserved) gene deserts. Twice as many gene deserts are variable, (non-conserved) and show very little evidence suggesting regulatory function.
Comment by KC — January 1, 2010 @ 6:37 pm
January 1st, 2010 at 6:39 pm
Steve,
Thanks but you and Nick should focus on supporting the claims of your position.
Also the HOX gene cluster was but one example.
And Meyer isn't limiting his discussion to genes.
Comment by ID guy — January 1, 2010 @ 6:39 pm
January 1st, 2010 at 7:51 pm
In one of America's largest cities an entrepreneur constructed a parking facility marvel. A modest plot of land was the spot of a spiraling multilevel series of lots whose area to parking space ratio was maximized to provide what the owner described as a wonder of parking density. On a cold winter day a blizzard hit town and dropped much snow. Some of it drifted through the spiraling ramps and was deposited inside the parking areas. Much of it melted inside the parking construct but at night the outside temperatures dropped to sub-zero levels and the moisture inside froze to ice.
The next morning the owner examined his lots. He found that 40% of the usable parking spaces had iced over and become unusable "junk space." His manager remarked that the wonder of parking density was no longer, strictly speaking, an accurate term. The owner retorted: "Call it what you want. This place is still one helluva design."
Genomes are one helluva design- Nick's gotcha argument notwithstanding.
Comment by Bradford — January 1, 2010 @ 7:51 pm
January 1st, 2010 at 8:19 pm
ID Guy writes that Nick and I need to "focus on supporting the claims of our position," in a thread that was specifically created to elicit critique of Meyer's book. I'm not fooled. I doubt Nick is fooled. Do you want to discuss genomes and their information "density," ID Guy? Because, friend, that's the topic here.
Bradford, Nick's "gotcha argument" is a much more serious problem than you seem to realize. To see why, you need to add a character to your nice parable. The character — we'll call him Steve — looks at all the ice and says to the owner, "Junk space? It's not junk at all! That ice is wonderful stuff, just wonderful. You couldn't park here without it." Without that character, your parable is a strawman. With that character, Nick's argument is a serious one, that Meyer and his defenders must deal with.
And in my view, your parable contains at least one key assumption, which Meyer seems to make also: that parasitic elements such as SINEs and LINEs are muck that has been added to an otherwise highly optimized system. Is that what you intend to assert?
Comment by SteveMatheson — January 1, 2010 @ 8:19 pm
January 1st, 2010 at 8:46 pm
Steve:
Not at all. Some transposable elements appear to be functional. Perhaps cells are able to turn "junk" into something functional or maybe what we perceive as junk is a misperception built on ignorance. Or maybe it really is junk. In any case an argument for design can be fashioned around genomic parts known to have function. An atelic perspective would assign junk as incidental to an atelic causal trail. But I would argue that non-functional genomic elements are indicators of suboptimal function (metabolically wasteful genomic stretches). However, not of an absence of design at point of origin. That's the point of the ice story.
Comment by Bradford — January 1, 2010 @ 8:46 pm
January 1st, 2010 at 8:57 pm
SteveMatheson says:
Meyer seems to make also: that parasitic elements such as SINEs and LINEs are muck that has been added to an otherwise highly optimized system. Is that what you intend to assert?
But Meyer says:
SINEs (Short Interspersed Nuclear Elements) are densest in housekeeping regions, for instance, where they modulate a host of genomic activities. LINEs (Long Interspersed Nuclear Elements) are by contrast, a common feature of tissue sepcific chromosome domains where they commonly act as chromosome "scaffold attachment regions" that determine how strands of DNA will fold and unfold in the nucleus. They also function as "molecular rheostats" that fine-tune gene expression.
Comment by Pez — January 1, 2010 @ 8:57 pm
January 1st, 2010 at 9:12 pm
Steve,
You are right- I was wrong.
My apologies.
Happy New Year
Comment by ID guy — January 1, 2010 @ 9:12 pm
January 1st, 2010 at 9:26 pm
For the record-
The "no junk/ limited junk DNA", although more likely in a design scenario should not be used because there really isn't any way of predicting how much "junk", real junk, can accumulate. Meaning if there was a lot of junk then that would not refute the design inference.
I would say that there mere fact there are regulators, transcription factors, switches, alternative gene splicing, proof-reading, error-correction and recycling is plenty of evidence for the design inference.
Moving on to Nick's complaint-
I know Meyer references the ENCODE project, and as Pez- thank you very much- pointed out it could be that there is a reason for SINEs and LINEs, transposons carry within their sequence the codes for the enzymes required to move them around, so their could be something to them- some reason we just haven't thought of.
Comment by ID guy — January 1, 2010 @ 9:26 pm
January 1st, 2010 at 10:10 pm
Trying this again.
Matzke's semantic argument quotes come from the epilogue of the book in which Meyer is discussing primarily the explanatory power of ID, not its predictions. And in the 17 page Appendix A: Some Predictions of Intelligent Design junk DNA takes up about 6 lines on the 12th page.
Comment by Pez — January 1, 2010 @ 10:10 pm
January 1st, 2010 at 11:43 pm
ID Guy, no worries. Happy New Year to you, too. How 'bout I come back when I've read the book and written my review. See you then.
Comment by SteveMatheson — January 1, 2010 @ 11:43 pm
January 2nd, 2010 at 12:59 am
Nick Matzke wrote a comment on his own article here. Quoting:
If you have not read Signature in the Cell rest assured that the 609 page book discusses much more than information density. What Nick is doing is combing a book in search of what he believes to be its weakest point and then directing his fire at it. He has not mentioned many other points but he does not have to. They will be brought out here. One way to assess a position is to reason what the author's main theme would be like if a disputed section were omitted. Would the author's case fall apart or would it still stand. In other words if you impute the most uncharitable interpretations to Meyer about information density and credit Matzke's arguments as completely plausible would that show that there are no signatures of design in cells? The answer is no. Imputing cellular design does not hinge on information density. Debunking design is not accomplished by assuming the correctness of Matzke's argument. Nick went fishing and pulled in a fish large enough to feed his followers. It's not big and powerful enough though to overcome the assumption tests just mentioned. We'll find out what Matzke et. al have to say in reponse the the totality of the book. We already know one thing. If you are unpersuaded by Nick's attitude toward ID, you no doubt are clueless.
Comment by Bradford — January 2, 2010 @ 12:59 am
January 2nd, 2010 at 1:59 am
Whoops!!! I accidentally moved a comment to the memory hole!!! How do we fix it?
Comment by Bilbo — January 2, 2010 @ 1:59 am
January 2nd, 2010 at 2:06 am
It has happened before Bilbo. The only thing you can do that I know of is to copy and paste the message from the hole and repost it under your name with an explanation preceding the comment.
Comment by Bradford — January 2, 2010 @ 2:06 am
January 2nd, 2010 at 2:20 am
statemachinist:
Thanks for that comment. Your comment was delayed in the queue for quite some time.
Comment by Bradford — January 2, 2010 @ 2:20 am
January 2nd, 2010 at 3:27 am
Mods — my original post in this thread seems to have disappeared? Weirdness…
My take on the thread this far: several ID advocates have basically conceded that my criticism is correct (although some of them are arguing that it doesn't matter), and the rest are attempting to save Meyer by hopeful parsing of his words to mean things that no reader would have gotten from a straight reading. This means I've made a pretty good point here.
(Although, it is true that it doesn't logically discredit Meyer's other arguments, it should make people suspicious of his competency to get his other biological facts, and his reasoning from them, correct.) And, this error I found it pretty simple and obvious. Why didn't the various Meyer fans out there find it? Thomas Nagel is a smart guy, for instance — why couldn't he see this simple self-contradiction. Was he was paying attention?
Replies to a few things:
What happens to the argument I presented in this thread? Nothing. The argument in this thread relies entirely on Meyer's arguments. Even if LINEs and SINEs have a function of some sort, it can't be a very "complex", sequence-specific, "information dense" one — because they are essentially the same sequence, repeated thousands or millions of times in our DNA! And Meyer himself has told us that such repetitive patterns are predictable results of simple algorithms, and therefore contain little complexity or information!
That said, even if you are just worried about "function" in some general, information-poor sense (Meyer explicitly connects specified complexity, i.e. coding, language-like, information-dense sequence, to function, throughout the entire book!!!! So good luck disconnecting those!!), keep in mind (a) the fact that function — or more commonly some kind of detectable biochemical effect — has been discovered for a LINE or SINE here or there in no way proves or even implies (b) the idea, extremely common with poorly-informed ID advocates, that ALL or almost all of these sequences must have a function. In fact, we have a lot of positive evidence AGAINST the idea that most LINEs and SINEs have some kind of sexy function — notably, (c) they are known to be the byproducts of retrotransposons and thus their presence has an easy, obvious, completely straightforward nonfunctional (for the organism) cause, and (d) the number of repetitive elements varies widely between very similar species, and doesn't appear to correlate with organismal complexity. E.g. some onions have genomes 5 times bigger than humans (mostly due to more repetitive elements), other onions have genomes 25 times bigger, and some fishes have genomes 10 times smaller than humans (but the same amount of genes, mostly the same developmental and hox genes, etc.).
These variations in DNA amount may have some kind of functional significance in bulk (one theory, much debated, is that large cells need large nuclei, and the amount of DNA physically determines the size of the nucleus) — but we pretty much know for sure that there isn't a lot of complex coding information in these variations.
Unfortunately, "we" apparently doesn't include Meyer, who apparently knows nothing about the well-known fact of genome size variability, despite the fact that it nukes the ID movement's junk DNA arguments…
PS: I and others have presented these facts many times before on TT and elsewhere, for the life of me I can't figure out why they are never absorbed and mentioned by IDists in their books and blogs. Instead they prefer to keep on keeping on, as if genome size variability didn't exist…
example refs:
http://pandasthumb.org/archive...
http://sandwalk.blogspot.com/2...
Comment by nickmatzke — January 2, 2010 @ 3:27 am
January 2nd, 2010 at 3:41 am
Oops, what I meant was that some onions have genomes 2 times as big as humans, and other onions in the same genus have genomes ~10 times as big as humans, or 5 times bigger than other onions.
Comment by nickmatzke — January 2, 2010 @ 3:41 am
January 2nd, 2010 at 11:29 am
You are a legend in your own mind.
To support your claim you linked to an out-of-date book.
Also it doesn't matter if the function of SINEs and LINEs is complex.
All that is required is that a function be present.
And what does genome size have to do with anything?
And perhaps IDists don't repeat what you post because we consider the source.
Sorry Nick but you have been so wrong about ID in the past it is difficult to take your word for anything.
Comment by ID guy — January 2, 2010 @ 11:29 am
January 2nd, 2010 at 12:34 pm
Nick:
That's probably the comment that was accidentally holed or deleted by Bilbo. I'll look at the rest of the comment later.
Comment by Bradford — January 2, 2010 @ 12:34 pm
January 2nd, 2010 at 12:48 pm
Sorry Nick, that was me. Trying to fix it. How do I get to the memory hole?
Comment by Bilbo — January 2, 2010 @ 12:48 pm
January 2nd, 2010 at 12:54 pm
From Nick's first link in his last post:
Who said anything about a "universal function"?
1- "this function" is in question
2- Just because we don't know now doesn't mean we will always be ignorant.
3- And basing an argument on ignorance, which is what Ryan Gregory is doing, is not the correct way of debating.
Comment by ID guy — January 2, 2010 @ 12:54 pm
January 2nd, 2010 at 12:59 pm
Nick's second link mentions the number of genes- the "G-Value paradox".
One way around that is alternative gene splicing (AGS).
With AGS one "gene" can become many genes.
That is why (in humans anyway) the number of proteins out numbers the number of genes.
Comment by ID guy — January 2, 2010 @ 12:59 pm
January 2nd, 2010 at 12:59 pm
No luck finding the memory hole, yet. Thank God I'm not a brain surgeon.
Comment by Bilbo — January 2, 2010 @ 12:59 pm
January 2nd, 2010 at 1:19 pm
SITC
page 459
Comment by Pez — January 2, 2010 @ 1:19 pm
January 2nd, 2010 at 1:28 pm
I think this is Nick's original comment. I don't know how to restore it to where it originally was.
Submitted on 2009/12/31 at 6:12pm
See, this is good old-fashioned TT fare here.
I have read Signature in the Cell. There are many problems with the argument, here is one. Meyer says that "information" — sequence-specific function — is densely concentrated in the DNA genome:
"Thus, far from being dispersed sparsely, haphazardly, and inefficiently within a sea of nonfunctional sequences (one that supposedly accumulated by mutation), functional genetic information is densely concentrated on the DNA molecule." (p. 461)
"Far from containing a preponderance of "junk" — nonprotein-coding regions that supposedly perform no function — the genome is dominated by sequences rich in functional information." (p. 461)
Furthermore, says Meyer, not only is this established truth, but it is a prediction of ID theory, and furthermore it was predicted by ID advocates a decade or more ago:
"The genome does display evidence of past viral insertions, deletions, transpositions, and the like, much as digital software copied again and again acumulates errors. Nevertheless, the vast majority of base sequences in the genome, and even the many sequences that do not code for proteins, serve essential biological functions. Genetic signal dwarfs noise, just as design advocates would expect and just as they predicted in the early 1990s." (p. 461)
However, at numerous places in the book, Meyer notes (correctly) that repetitive sequences have little information:
"Since information and improbability are inversely related, high-probability repeating sequences like ABCABCABCABCABCABC have very little information (either carrying capacity or content). And this makes sense too. Once you have seen the first triad of ABCs, the rest are "redundant"; they convey nothing new. They aren't informative. Such sequences aren't complex either. Why? A short algorithm or set of commands could easily generate a long sequence of repeating ABCs, making the sequence compressible." (p. 107)
Unfortunately for Meyer, he seems to not realize that 40-50% of the human genome (and most animal genomes of similar size) consists of LINEs, SINEs, segmental duplications, and other repeating elements. As documented here:
http://www.ncbi.nlm.nih.gov/bo...
In other words, there is no way that in "the vast majority" of the genome genetic information is "densely concentrated" — as proven by his own arguments!
QED.
Comment by Bilbo — January 2, 2010 @ 1:28 pm
January 2nd, 2010 at 1:55 pm
Specified complexity is complex sequencing which enables biological function But where in the book does he claim that an entire genome must be characterized by specified complexity?
Nick quoting Meyer:
A gene can be described as containing densely concentrated functional genetic information correct?
Comment by Bradford — January 2, 2010 @ 1:55 pm
January 2nd, 2010 at 2:18 pm
The information density Meyer is talking about is not merely a simplistic measure of the linear sequences. As emphasized above, he is talking about functional information; codes within codes, messages within messages, information upon information. He is talking about how much information is communicated, in various ways, from the DNA molecule to the organism.
Comment by Pez — January 2, 2010 @ 2:18 pm
January 2nd, 2010 at 2:32 pm
Is there a formal definition of information density?
Comment by Bradford — January 2, 2010 @ 2:32 pm
January 2nd, 2010 at 2:58 pm
Yes, that's it.
Comment by Mung — January 2, 2010 @ 2:58 pm
January 2nd, 2010 at 3:21 pm
I doubt it.
97
Comment by Pez — January 2, 2010 @ 3:21 pm
January 2nd, 2010 at 3:22 pm
Bradford, I'm a bit confused by your comment above. Are you asserting that Meyer's claim (the one that begins "Far from being dispersed sparsely…") is simply a claim that when "functional genetic information" is found in a genome that it will be "dense"?
I ask because your challenge to Nick ("where in the book does he claim that an entire genome must be characterized by specified complexity?") seems to be answered in the very quotation that follows your challenge.
Bradford, I don't think Meyer is stupid, and he'd have to be stupid or dishonest to claim that the entire human genome is characterized by dense deposition of "functional genetic information." But Nick has shown that Meyer makes statements that are confusing at best, misleading at worst. Whatever Meyer means by "a sea of nonfunctional sequences," it remains a fact that the human genome (as you have already acknowledged) is so riddled with parasitic elements that they are thought to compose around half of its content. And that's just the human genome, which is practically skeletal compared to the giant genomes of some other organisms. If Meyer wrote a book on design in the genome without dealing specifically with such phenomena, then he never intended to seriously discuss genomic structure and origins. And that's a huge problem, guys. Huge.
Comment by SteveMatheson — January 2, 2010 @ 3:22 pm
January 2nd, 2010 at 3:27 pm
Meyer asserts that there is "orders of magnitude" more information in the genome than previously thought, but ignores the e.g. fact that there are far fewer genes than many people used to think (23,000 instead of 100,000). But more importantly, "orders of magnitude" implies that Meyer has a rigorous definition, and quantitative measurements of the amount of information in the genome — but he presents neither of these.
Re: information and specified complexity: for Meyer these are the same:
"The term specified complexity is, therefore, a synonym for specified information or information content." (p. 107)
Finally, more problems for the ridiculous "he only meant parts of the DNA have dense information" interpretation:
Comment by nickmatzke — January 2, 2010 @ 3:27 pm
January 2nd, 2010 at 3:36 pm
This is not ignored; it is the point. A gene count is not an accurate measure of the information in the genome.
Comment by Pez — January 2, 2010 @ 3:36 pm
January 2nd, 2010 at 3:44 pm
We have already covered that Nick- alternative gene splicing can turn 23,000 into a much larger number.
Also there are "orders of magnitude" more information than previously thought.
ENCODE has pretty much confirmed that.
Meyer is not limiting his claim to genes.
He is talking about the meta-information- that which controls the genes- the information your position cannot explain.
However it has become clear that Nick's "argument" is a semantic one.
And as I said before if you really wanted to show that Meyer was wrong all you would have to do is to actually support the claims of your position.
That way you wouldn't have to rely solely on your interpretation of what Meyer is saying.
Comment by ID guy — January 2, 2010 @ 3:44 pm
January 2nd, 2010 at 4:00 pm
Steve:
I hesitate to mold a discussion around the term dense unless there is a formal definition for it. If there is no rigorous definition of dense then it can mean whatever someone thinks it means and you know where that goes. It would have been better if Meyer had not used the term and instead stuck with specified complexity which ties complexity to biological function. That raises issues which are problematic for ID critics. I think you, me and Meyer could agree that protein coding genes are information rich. Short repetitive functional sequences in intergenic regions are not information rich. Does that sound reasonable to you?
If Meyer believes that entire genomes are specifiably complex then I have a problem with that. But I do not think that is his position. In the follow up quote Meyer said that "functional genetic information is densely concentrated on the DNA molecule." That's not the same as saying that the entire genome must be characterized by specified complexity.
Comment by Bradford — January 2, 2010 @ 4:00 pm
January 2nd, 2010 at 4:02 pm
Is New York City densely populated?
Comment by Pez — January 2, 2010 @ 4:02 pm
January 2nd, 2010 at 4:08 pm
Hey Bradford, thanks for the clarification. We are on the same page, and that's good to see. It will be fruitful to reconnect when I've read and reviewed the book.
Comment by SteveMatheson — January 2, 2010 @ 4:08 pm
January 2nd, 2010 at 4:13 pm
Where in the book are you taking the orders of magnitude from? On the issue of number of genes and prior expectations, it does not strengthen your critique to emphasize that fewer genes than expected were found without simultaneously pointing out that multiple proteins coded for by the same gene was at one time also unexpected as was the regulatory role found to be linked to genes coding for truncated RNA.
Comment by Bradford — January 2, 2010 @ 4:13 pm
January 2nd, 2010 at 4:20 pm
Hi Bradford,
He took it from my quote which first discusses the recently discovered layers of information:
Comment by Pez — January 2, 2010 @ 4:20 pm
January 2nd, 2010 at 4:20 pm
Steve:
I'm looking forward to that. I hope I can post a blog entry on your blog write up of Meyer here. In the meantime I'll likely post some more on Meyer emphasizing other points.
BTW, someone whose views I value, gave a good personal reference about you in an off-list message.
Comment by Bradford — January 2, 2010 @ 4:20 pm
January 2nd, 2010 at 4:21 pm
Thanks for that Pez. Good to see you again.
Comment by Bradford — January 2, 2010 @ 4:21 pm
January 2nd, 2010 at 4:31 pm
Thanks, Bradford.
Happy New Year.
Comment by Pez — January 2, 2010 @ 4:31 pm
January 3rd, 2010 at 12:22 am
WRT the challenge in the OP, I'll mention one minor subject on which Meyer got his facts wrong, namely his depiction of Tom Schneider's ev in Chapter 13.
Meyer says that Schneider "indicates that the Ev program has thus created 131 bits of information." Schneider didn't actually say that, but Meyer tells us in footnote 35 how he (Meyer) infers it from Schneider's paper:
Schneider indicates that his computer organism has 265 digital characters, roughly 134 or so of which are dedicated to computing. That leaves 131 digital characters corresponding to the binding site, the generation of which Ev attempts to simulate (see the caption to Figure 1,
p. 2796). A successful simulation would, therefore, involve specifying 131 digital characters ("zeros" and “ones”), or 131 bits of information.
But if we read the caption to which Meyer refers, we see that the correct numbers are not 265, "roughly 134 or so", and 131, but rather 256, 125, and 256. The erroneous 131, as well as the "zeros" and "ones", come from Marks and Dembski's analysis, not from Schneider's paper. The erroneous 265 and "roughly 134 or so" seem to be Meyer's own mistakes.
As Meyer describes ev, he talks about a filter that is applied before selection takes place:
Alter favoring sequences that manifest the general profile of a binding site, Ev applies a fitness function to the remaining sequences.
and
Ev incorporates one additional step that Dawkins’s and Kiippers’s simulations lack. Before Ev applies its fitness function, it applies a filter to the crop of mutated sequences. The filter favors sequences that have the general profile of a binding site. Like the fitness function, this coarser filter makes use of information about the functional requirements of binding sites to favor some sequences over others. As such, it imparts information based on knowledge that Thomas Schneider, not natural selection or the environment, has imparted into the Ev simulation. Ev exhibits the genius of its designer.
This is wrong in several ways. First, there is no pre-selection filter — selection is applied to every member of the population. Meyer is probably thinking of the function that maps the genetic sequence to positions of the binding sites. Because of the nature of this function, some sets of binding site positions are easier to generate than others. The set that Schneider chose is easier to generate than others, but there is no evidence that Schneider knew this, and generating Schneider's set is still completely intractable without the mutation+selection mechanism.
Further on, Meyer says:
Marks shows that each time the program uses knowledge of the target sequence to exclude some sequences and preserve others, it imparts a quantifiable amount of information in its selection. Marks quantifies these sources of active information and shows that they reduce the difficulty well below the 131 bits Schneider claims that Ev can produce "from scratch."
Again, Schneider never said anything about producing 131 bits from scratch. And Meyer is correct that Marks quantified the sources of active information, but those quantities turned out to be wrong by tens of orders of magnitude, and the paper was withdrawn and rewritten with a non-quantitative analysis. This occurred about two years before Meyer published TSITC.
Comment by R0b — January 3, 2010 @ 12:22 am
January 3rd, 2010 at 8:06 am
R0b, do you believe Ev is useful and if so useful for what purpose?
Happy New Year to you too Pez.
Comment by Bradford — January 3, 2010 @ 8:06 am
January 3rd, 2010 at 1:53 pm
"riddled with parasitic elements"? "practically skeletal"? This type of hyperbole indicates to me that you've already taken a stand on the issue without having read the book.
Let me ask you this though: These "parasitic elements", have they degraded the genome from its original form? And does "skeletal" mean the genome was once "full-bodied"?
Comment by Daniel Smith — January 3, 2010 @ 1:53 pm
January 3rd, 2010 at 2:46 pm
It finally dawned on me that I could just paste Nick's first post in with his second post. D'oh!
Comment by Bilbo — January 3, 2010 @ 2:46 pm
January 3rd, 2010 at 7:40 pm
Daniel, the human genome is riddled with parasitic elements, as Bradford wisely acknowledges. (So, it would seem, does Meyer.) My understanding of this feature of the human genome has nothing to do with Meyer's book. What you call "hyperbole" is a consensus view of molecular geneticists and those who understand genomes, and if you have the impression that I've already reached certain conclusions about genomic structure, you're right: I'm a professional biologist. My point, and Nick's as well, is not that Meyer is wrong about design of the genome, but that if he (or you or anyone else) is asserting that the human genome is a densely-packed, highly-optimized bundle of superpower, then he (and you) have to deal with the 40+% of the typical eukaryotic genome that is composed of known parasitic (and often mobile) genetic elements.
And if you'd read my comment more carefully, you'd see that by "skeletal" I meant "small." The human genome is dwarfed in size by many other animal (and plant) genomes. Only a fool would propose a serious explanation of genomic structure without seeking to account for that fact.
Comment by SteveMatheson — January 3, 2010 @ 7:40 pm
January 3rd, 2010 at 8:49 pm
[...] Matheson and others have discussed mobile genetic elements in The Signature in the Cell Challenge. But we need not presume that retrotransposons necessarily correlate to functionless sequences. They [...]
Pingback by Retrotransposon transcription - Telic Thoughts — January 3, 2010 @ 8:49 pm
January 3rd, 2010 at 11:24 pm
Is there now an admission by biologists that at least 50% of the human genome has function?
Comment by Bilbo — January 3, 2010 @ 11:24 pm
January 4th, 2010 at 12:05 am
Dr Hubbard said: "We are now seeing the majority of the rest of the genome is active to some extent."
…
Dr Birney added that many of the RNA molecules were copying overlapping sequences of DNA.
He said: "The genome looks like it is far more of a network of RNA transcripts that are all collaborating together. Some go off and make proteins; [and] quite a few, although we know they are there, we really do not have a good understanding of what they do.
http://news.bbc.co.uk/2/hi/sci...
Comment by Pez — January 4, 2010 @ 12:05 am
January 4th, 2010 at 12:38 am
The not-true Scotsman, biologist Sternberg and junk DNA.
http://www.richardsternberg.or...
Search "McClintock" (98% functional) and "DNA" (the so-called excess and non-coding sequences in genomes are functional, nay multifunctional, and thus they contain codes and multiple levels.)
Comment by Pez — January 4, 2010 @ 12:38 am
January 4th, 2010 at 2:01 pm
Bradford:
If anyone is surprised by the results of programs like ev, Avida, or even Dawkins' Weasel, then I suppose the programs are useful in that they cause those people to question their intuitions regarding complex adaptive systems. But my layman's hunch is that they probably don't give us any non-trivial insights into biology. I would guess that the non-linearity and positive and negative feedback loops of biological processes render computer models unreliable predictors of real-life behavior.
I realize that some people put forth ev and the like as counterexamples to fundamental claims made by some ID proponents, e.g. "Chance and necessity cannot create information", or "Mutation+selection cannot generate new functional features". My opinion is that such claims are usually poorly-defined, and that it's premature to come up with counterexamples before fleshing out exactly what the claims are saying. And I think that the responses by Marks, Dembski, and Meyer only exacerbate the communication problem, and we end up with the two sides talking past each other. That's my 2 cents, and worth every penny except for two.
BTW, I apologize for the strange characters in my comment above, which show up in IE but not in Chrome. Side effects of using OCR to copy text from Kindle for PC.
Comment by R0b — January 4, 2010 @ 2:01 pm
January 4th, 2010 at 4:54 pm
Bilbo asked, "Is there now an admission by biologists that at least 50% of the human genome has function?" I assume he was responding to my comment above that referred to the "40+% of the typical eukaryotic genome that is composed of known parasitic (and often mobile) genetic elements."
The answer is no. First, I was referring specifically to the 40-odd percent of the human genome made up of various parasitic elements. There are other genomic elements that are likely to be largely dispensable, most notably introns, which make up about a quarter of the human genome. Second, there is no consensus among biologists about how much function to expect from these large classes of genomic elements that don't have identifiable and characteristic functions. Opinions on this matter vary widely. Again, I recommend Ryan Gregory as a clear source on such questions.
Comment by SteveMatheson — January 4, 2010 @ 4:54 pm
January 4th, 2010 at 5:09 pm
http://www.sciencedaily.com/re...
Comment by Pez — January 4, 2010 @ 5:09 pm
January 4th, 2010 at 8:58 pm
I didn't realize you were a professional biologist. I also know though, that that doesn't automatically make you right.
You didn't really answer my questions: Have these "parasitic elements" degraded the genome from its original form?
And does "skeletal" (which you've now defined as "small") mean the genome was once much larger?
Comment by Daniel Smith — January 4, 2010 @ 8:58 pm
January 4th, 2010 at 9:33 pm
I don't know what your mean by "its original form." But I'd say that a characterization of the accumulation of parasitic elements as "degradation" is quite simplistic. For one thing, most mobile elements alter gene expression locally and can thereby induce phenotypic changes. If those changes lead to increased fitness, then you have a parasitic element whose arrival has an interesting cost/benefit curve. And we know that some retroviral elements seem to have been co-opted by the genome in the formation of new genes.
So it seems that parasitic elements can be destructive (typically) or roughly neutral (not as typically) or beneficial (rarely).
Interesting that you used scare quotes around parasitic elements. Perhaps you dispute the use of the term parasite to describe a virus?
The human genome has not changed dramatically in size, as far as I know, but again I don't know what you mean by your question. Genomes vary a lot in size among organisms, and the human genome is dwarfed by many other genomes. This difference in size is due almost completely to mobile parasitic elements. Some species of vertebrates, for example, have comparatively small genomes compared to humans, while many species of animal and plant have much larger genomes than humans. And the differences have nothing to do with the number of genes. They are accounted for by the numbers of parasitic elements.
So for example, birds and flying mammals have somewhat small genomes — smaller by hundreds of millions of base pairs than genomes from similar non-flying vertebrate species. This could mean that the common ancestor had a small genome, and the non-flying species accumulated bloat by letting parasitic elements run wild. Or it could mean that the common ancestor had a typically large genome with the usual collection of parasitic debris, but then the flying species put the kabosh on that and restricted the further accumulation of mobile elements. Or it could be a bit of both.
Daniel, my point is that a theory that claims to identify design in the genome needs to account for such observations carefully. That's all. Just think about it.
Comment by SteveMatheson — January 4, 2010 @ 9:33 pm
January 5th, 2010 at 1:51 am
This has been a really interesting thread. Thanks to Pez for repeatedly getting back to the actual text, and to statemachinist as well.
It is interesting that a warning is made that the thrust of Meyer's book should be questioned over such things as the use and placement of a descriptive word like "vast" etc. And yet, those who have made that argument have had their own fact statements (not just adjectives) contradicted by updated information linked to on this very thread.
Comment by Upright BiPed — January 5, 2010 @ 1:51 am
January 7th, 2010 at 1:18 pm
Some intron serendipity.
http://www.evolutionnews.org/2...
Comment by Pez — January 7, 2010 @ 1:18 pm