More Recent Comments

Wednesday, March 27, 2013

ENCODE, Junk DNA, and Intelligent Design Creationism

andyjones has replied to my earlier posting on ENCODE and junk DNA. You can read his response at: (More) Function, the evolution-free gospel of ENCODE. Here's part of what he says ...
Larry Moran has sort-of replied to my previous blogpost but disappoints with only one substantive point. And even that one point is wrong: ID is not committed to the idea that individual genomes be well-designed; that is just an expectation some of us derive based on belief in a designer which is established on other evidence. ID would still be true if only globular proteins were designed (lookup Axe), or even if only the flagellum was designed (lookup Behe), or even if only the first life form was designed (lookup Meyer – and please read their actual work, not cheap reviews, because reviewers often dont pick up on the salient points – more below). I just say this lest readers get the impression that this is ID’s strongest point, or in any sense a weak point. It is neither.
It's true that there are some IDiots who are distancing themselves from a commitment to junk DNA. There are probably some who claim that they could live with the fact that 90% of our DNA is junk.

But let's not forget that Jonathan Wells is a prominent IDiot and he wrote a book on The Myth of Junk DNA. It sounded very much like Intelligent Design Creationism is staking its reputation on finding function for most of our genome.

Have you heard of someone named Stephen Meyer? He wrote a book called Signature in the Cell and that book seems to be widely admired in the IDiot community. I think it's been mentioned once or twice on the Intelligent Design Creationist blogs. I blogged about what Meyer wrote about junk DNA a few years ago [Stephen Meyer Talks About Junk DNA]. Here's part of what's in that post, quoting from Stephen Meyer's book ....
ID advocates advance a different view of nonprotein-coding DNA. The theory of intelligent design predicts that most of the nonprotein-coding sequences in the genome should perform some biological function, even if they do not direct protein synthesis. ID theorists do not deny that mutational processes might have degraded or "broken" some previously functional DNA, but we predict that the functional DNA (the signal) should dwarf the nonfunctional DNA (the noise), and not the reverse. As William Dembski explained and predicted in 1998: "On an evolutionary view we expect a lot of useless DNA. If, on the other hand, organisms are designed, we expect DNA, as much as possible, to exhibit function." The discovery in recent years that non-protein coding DNA performs a diversity of important biological functions has confirmed this prediction. It also decisively refutes prominent critics of intelligent design—including Shermer Miller and Kitchener—who have continued to argue (each as recently as 2008) that the genome is composed of mostly useless DNA.
Then there's the famous Casey Luskin who is never one to back away from a prediction. Here's what he wrote in Intelligent Design and the Death of the "Junk-DNA" Neo-Darwinian Paradigm.
Intelligent Design has Long Predicted This Day

Proponents of intelligent design have long maintained that Neo-Darwinism's widely held assumption that our cells contain much genetic "junk" is both dangerous to the progress of science and wrong. As I explain here, design theorists recognize that "Intelligent agents typically create functional things," and thus Jonathan Wells has suggested, "From an ID perspective, however, it is extremely unlikely that an organism would expend its resources on preserving and transmitting so much ‘junk'." [4] Design theorists have thus been predicting the death of the junk-DNA paradigm for many years:
I guess these guys (Wells, Meyer, Dembski, and Luskin) don't speak for andyjones but I think it's fair to say that they represent the consensus view of the average IDiot. Let's stop pretending that junk DNA isn't an affront to intelligent design. Junk DNA is a major threat to their position and that's why they're so concerned.

Let's look at a bit more of what andyjones has to say ..
In case you missed it, the interesting thing about ENCODE is the discovery that 80% of the human genome is transcribed.
That was somewhat "interesting" in 2007 when the ENCODE pilot project was published but it wasn't the least bit interesting last September. Scientists have known since the mid-1970s that a large fraction of our genome is complementary to RNA in the cell. We've known for almost as long that most "pervasive" transcription is very likely useless transcription or accidental transcription. This is consistent with a lot of papers showing that RNA polymerase binds at many sites besides functional promoters. It's in most of the textbooks. This explanation became more widely known following publication of the pilot study in 2007. Unfortunately, it's not only the IDiots who ignore it but also many of the ENCODE workers.
Darwinian theories did not predict this.
That's correct. Darwinian theory is focused on natural selection as the most important mechanism of evolution. Strict Darwinists predicted that there would be very little junk in our genome. Most of them still stand by that prediction and that's why there are so many ENCODE workers who are opposed to junk DNA. (The IDiots know this because I've recently been assured that they understand the difference between Darwinism and modern evolutionary theory.)
For anyone who does not have a strong upstream thought-filter*, and has not been brain-washed by the cult, transcription is strongly indicative of function. Thus ENCODE’s headline is on target.
Yes, it's true that most—but not all—functional regions are transcribed and it's true that transcription can be indicative of function, especially if the transcripts are abundant.

On the other hand, regions of the genome that are transcribed very rarely do not turn out to be reliable indicators of function because we have a better explanation. Most of pervasive transcription falls into this very rare category.
It is quite reasonable now to expect that details of actual function will subsequently be found for much of the genome. Therefore we should keep looking for that function.
Scientists have accumulated plenty of evidence that most of our genome is not functional in any meaningful sense of the word. It's unlikely that this evidence will be overturned or discredited in the near future and it's unlikely that we will find functions for any significant portion of our genome. There are some very good scientists who have been looking for decades.
This site (UncommonDescent) has many articles on elements of DNA, previously thought junk, that turned out to be of importance. For examples, pseudogene transcripts regulate those of real genes, and ERV-elements turn out to be important in developing embryos.
It's true that Uncommon Descent has been a bit obsessive about the issue. It almost looks like discrediting junk DNA is a big deal for the IDiots in spite of what andyjones says above. If we add up all of the functional regions of the human genome that have appeared on the blog—or in Jonathan Wells' book— they probably amount to about 0.1% of the genome ... and that's being generous.
As I said, quite reasonably I think, ‘Time will tell’ for all the rest as well.
At the rate they're going (0.1% in ten years) it's going to take about 9000 years to find new functions for the 90% of the genome we think is junk.

Most of the IDiots are going to be raptured long before that ....


Piotr Gąsiorowski said...

So, according to "andyjones", ID would still stand if anything whatsoever were designed, even granted that the rest weren't. That looks like an attempt to shift the burden of proof to the other side. Why commit yourself to potentially falsifiable claims ("X can't have evolved, therefore it's designed"), if instead you can say, "All right, perhaps X is not designed, but something else might be, so either you prove non-design for each and every structure, or I win."

Joe Felsenstein said...

The issue of junk DNA also exposes a contradiction in the ID position. Typically, when people make "bad design" arguments against ID, the ID types respond that one cannot judge the motives of the designer and one cannot presume to know whether any design will be good or bad, and one certainly cannot predict that designs will be good.

Then, when Junk DNA comes up, they blatantly contradict themselves. They are made very unhappy by the possibility that the genome is cluttered with junk, and they furiously seek to discredit the notion. So they really do believe in "good design" after all, in spite of their rejection of it as a test of ID.

On the other side evolutionary biologists had no prior commitment to the existence of Junk DNA before people started to observe an excess of DNA and the presence of nonsensical repeated sequences, later transposons, and later introns. The idea of junk in the genome isn't built in to evolutionary theory -- it got there by observations.

Diogenes said...

Joe: "Typically, when people make "bad design" arguments against ID, the ID types respond that one cannot judge the motives of the designer and one cannot presume to know whether any design will be good or bad, and one certainly cannot predict that designs will be good."

This leads them into a forest of self-contradictions.

Can you find the contradiction in these quotes below?

Michael Behe, refuting Ken Miller: "Arguments based on perceived faults or vestigial genes and organs run the danger of the argument of Diogenes [note: not me, the other Diogenes] that the progression of seasons shows intelligent design. It is scientifically unsound to make any assumptions of the way things ought to be." [Michael Behe, Darwin's Black Box (1996) p.227]

Above, Behe sniffs that Ken Miller is "scientifically unsound." Miller's "bad design" argument makes "assumptions of the way things ought to be", and is thus religious, not scientific.

Now compare Klinghitler:

David Klinghitler: "But ID's explanation [for why plants taste good] is more satisfying because it doesn't rely on luck [= natural selection]. In a view of nature informed by the theory of intelligent design, it seems reasonable to think that plants taste good... because that is part of a vision of how things should be, a vision proceeding from a purposeful plan." [Why There Have to Be "Pesky" Bugs. David Klinghoffer. ENV. October 5, 2012.]

According to Behe, ID is true science, not religion, precisely because it makes NO "assumptions of the way things ought to be."

According to Klinghitler, ID is superior science, better science, precisely because it "a vision of how things should be, a vision proceeding from a purposeful plan."

What Klinghitler presents as ID's main selling point, is precisely what Behe says makes an argument religious, not scientific.

Diogenes said...

The IDologues' magic puff of smoke allows their deities/ genies/ sorcerers to intelligently design non-functional stuff, if that's what they want to do.

So Casey Luskin, Esquire, in predicting ZERO Junk DNA, must be making metaphysical assumptions about the purposes of the intelligent designer. Doesn't he know that asserting to know the purposes of God is theology, not science? I learned that from... Casey Luskin, Michael Behe, and every Intelligent Design proponent.

If their deities /genies CAN make the vestigial wing on the kiwi, pseudogenes, percieved faults, etc., how do Behe and Luskin know there's no junk DNA? Is it still "scientifically unsound to make any assumptions of the way things ought to be", as Behe sniffs to Ken Miller?

Or is it only "scientifically unsound" when WE do it, and A-OK when you IDiots do it?

Phillip Johnson, founder of the Discovery Institute: " seems to me that the peacock and peahen are just the kind of creatures a whimsical Creator might favor, but that an "uncaring mechanical process" like natural selection would never permit to develop." [Phillip Johnson, Darwin on Trial, (1993), p. 31]

So if you IDiots can divine the mind of God, and know your deity would LIKE to make the NON-FUNCTIONAL stuff like the tail on the peacock, then how do you know for sure your deity didn't make tons and tons of non-functional DNA?

Casey Luskin, Esquire, sniffs: "You are making a theological argument, not a scientific one," [Luskin at ENV, 2012]

Casey Luskin, Esquire: "..."undesirable design" arguments are theologically based, whereas ID is NOT! [capitalization in original]

As a science, ID doesn't address theological questions about whether the design is "desirable," "undesirable," "perfect," or "imperfect." Undesirable design is still design..." [Luskin at ENV, 2012]

Ohhh, I see. Undesirable is OK, Junk is not OK. Gotcha!

Hold on Casey, how do you know your deities /genies /fairies didn't make tons and tons of Junk-- I mean uh, "undesirable"-- DNA?

Luskin again: "Holding biological systems to some vague standard of "perfect design"... is the wrong way to test ID."

Because it doesn't give the answer you want. Unlike all the other ways of testing ID, such as... such as... um... it'll come to me later, but let's continue with lawyer Luskin:

"Examples like broken machinery, computer failures, and decaying buildings all show that a structure might be designed by an intelligent agent... [Casey Luskin, "Undesirable Intelligent Design is Still Intelligent Design.", ENV, Feb 29, 2012]

I see! Broken machinery can be designed by an intelligent agent, but broken genes cannot be; broken computers can be designed by an intelligent agent, but broken transposons cannot be. Now that makes perfect sense.

But... if ID doesn't require optimal design, how do you know your deities /genies /fairies didn't make lots of non-functional DNA?

Oh I know, ooga booga. You're receeeiving a messssaggee... from the ooootherrrr siiiiide.... whooo... Oooooga Booooga....

The spirits' message is... repeat after meee... "Owha... Tajer... Kyam... Owha... Tajer... Kyam..." Repeat this again and again, faster and faster, until it reveals the innermost secret of your soul...

Anonymous said...

Venter's mycobacterium chromosome was modified from its natural counterpart and added some designed sequences, therefore ID stands!

Claudiu Bandea said...


You are probably the top blogger covering junk DNA (jDNA), so I want to ask you a direct question on this issue.

You might know that ever since the ENCODE fiasco started, I have tried to promote an old hypothesis on the evolution of genome size, c-value pardox, and the function of jDNA. According to this hypothesis, the so called jDNA provides a protective mechanism against insertional mutagenesis. For almost a quarter century since I publish it, I never tried to promote it, or write endless articles about it as often is the case in science; I thought that scientists will evaluate it, but that did not happen.

However, incited by ENCODE discussions, I decided to bring this old hypothesis forward. As I was joking in recent response to Joe Felsenstein, after a few months of comments, I finally convinced one person, that was him, to openly agree that jDNA provides a protective mechanism against insertional mutagenesis.

So here is my question: does jDNA serve as a protective mechanism against insertional mutagenesis or not? My answer is yes. What is yours?

John Harshman said...

I'm pretty sure you didn't convince Joe. I think he allowed that there might be a possibility of such a thing under certain conditions. But you have yet to deal with the C-value paradox or, to put it otherwise, the onion test. Do some species of onion have way better protection from insertional mutagenesis than others? If so, why would that be?

Joe Felsenstein said...

jDNA provides a protective mechanism (*if* it does not also more than offset this by increasing the rate of insertion, and I did say that). But I also said clearly, and Claudiu did not acknowledge this, that the effect of this protection in favoring additions of new pieces of jDNA are likely to be so small that selection based on it would be a negligible force. Simple parasitic insertion of new jDNA should be a much more important force in bringing about jDNA.

I urged Claudiu to make a quantitative analysis of this issue and deal with this criticism quantitatively before he bothers us again with his favorite hypothesis. But here he is again claiming I was converted to his view ...

Faizal Ali said...

So Claudiu is not just a crank, but a liar. All the more reason we should just return to ignoring his posts.

Larry Moran said...

All the more reason we should just return to ignoring his posts.

Please, please, please, ignore his comments so I don't have to ban him.

Claudiu Bandea said...


Here is what I wrote in our previous discussion (

“… just like your response here, your previous comment allowed for some “quite small” or “extremely weak selection” on retaining “a piece of junk DNA.” I completely agree with that: indeed, the selection acting on each piece of jDNA is extremely small”

And this is what you responded to that:

Good to see that you agree with me on that. That's progress

Why are you saying now, only a few weeks later, that I did not acknowledge it:

“But I also said clearly, and Claudiu did not acknowledge this, that the effect of this protection in favoring additions of new pieces of jDNA are likely to be so small that selection based on it would be a negligible force”

I think, you probably forgot what you wrote, but you need to be more careful because people like lutesuite (see below) can’t wait to use all kinds of derogatory language. [BTW, a few months ago, I suggested that Larry should not allow anonymous commenting on his blog, and he did recently ban it. I think it might be also a good idea to also ban the use derogatory language; I think we all know that anyone can get multiple blogging ‘identities’ and just use this type of language left and right).

In regard to your answer to my question whether jDNA provides a protective function against insertion mutagenesis or not, I said:

“I appreciate your YES,even with an ‘if’ attach with it”

So, there are not misunderstandings on my part.

Also, you say that I need to do “quantitative analysis of this issue”. Obviously, any model needs further developments, but when the so called jDNA constitutes 90% or the human genome, you don’t need quantitative analysis to prove its protective function against insertional mutagenesis, particularly against insertion of retroviruses, such as HIV, in somatic cells, which can lead to cancer.

Claudiu Bandea said...

John Harshman: Do some species of onion have way better protection from insertional mutagenesis than others? If so, why would that be?

Yes, they do. Clearly, a species of onion (O1) which has twice as much jDNA as another onion species (O2) increases its protection against insertional mutagenesis by an exogenous virus approximately 100%, or even more if there are additional mechanisms for directing the insertion elements preferentially in the jDNA.

Expectedly, as an adaptive defense mechanism, the amount of protective DNA varies from one species to another based on the insertional mutagenesis activity and evolutionary constrains on genome size.

AllanMiller said...

Looks like my theory of Probability Waves stands firm on similar grounds - some things are normally impossible, but once in a while a wave perturbs the probability towards one, and something Very Unlikely happens. I need only one Very Unlikely thing to happen by this means, and I win. I await proof that every unlikely thing ever was caused by something else.

Unknown said...

Larry, thanks for your efforts to bring honesty to the ID controversy. I have been a long-time reader here. In case it is of interest, I wrote a fairly detailed rebuttal of ID claims around ENCODE and junk DNA, drawing on your work among others. It's called "Junk_DNA_Design" on Letters To Creationists blog - can Google it.

(Had problems getting this comment to post, apologize for dups)
-Scott Buchanan

John Harshman said...

Not very satisfying. Your claim is that genome size varies adaptively based on insertional mutagenesis activity. But we would see exactly the same thing if genome size were merely the result of insertional mutagenesis activity. Your protection and the thing it's supposed to protect from are, in other words, the same thing. And you need to explain why two species of onion (and species in general, but most especially the closely related ones) need such radically different amounts of protection. And of course, you need to explain what you mean by evolutionary constraints on genome size.

Anonymous said...

Since the discussion seem to have averted, could someone pleeeease explain to me, why mycoplasama's genetic code would not work in human cells or insert a link to an easy to understan explanation?

I'm referring to the Ventor's video where he says that.

TheOtherJim said...

Strictly limiting to the code differences, there is a change in the genetic code, which will mess up the proteins encoded on the Mycoplasma DNA, if there is an attempt to decode it with the human cellular machinery.

In humans, the UGA triplet in an mRNA means "stop making protein here". In Mycoplasma, it means "insert tryptophan here". So all proteins translated from Mycoplasma-sequence RNA that tried to use UGA as Trp, would prematurely truncate (as the human system sees this as Stop instead).

Animal mitochondria have the same variation in the genetic code and a couple others (but are derived independently). The genes on our mitochondria therefore can only be translated correctly inside the mitochondria, and not by the system used by the rest of our DNA.

Claudiu Bandea said...

John Harshman:

Your protection and the thing it's supposed to protect from are, in other words, the same thing

As I said before: it’s like fighting fire with fire. This is somewhat similar to the antiviral defense system called CRISPR, in which the cells acquire viral sequences in their genome in order to use them as a defense mechanism against future viral infections.

Your claim is that genome size varies adaptively based on insertional mutagenesis activity

My main claim is that ‘jDNA’ provides a protective mechanism against insertional mutagenesis. As it is now well know, evolutionarily, jDNA originates from the activity of viral elements, therefore, it does not originate in order to protect the host genome from insertional mutagenesis. However, jDNA is maintained by natural selection to provide a protective function particularly against insertion by retroviruses in somatic cells, which can lead to cancer. How much of it is maintained depends of evolutionary constraints on genome size; for example, hummingbirds keep much less jDNA than humans or onions.

But we would see exactly the same thing if genome size were merely the result of insertional mutagenesis activity

I don’t understand exactly what you mean, can you elaborate?

Diogenes said...


You are probably the top blogger covering junk DNA (jDNA)

Ahem, cough! I'm way way waaay down on the totem pole! You should worry more about convincing bloggers more knowledgeable than myself, like Larry, Joe Felsenstein, T. Ryan Gregory, Michael Eisen etc., rather than myself.

You've failed with Larry and Joe, so that leaves Gregory and Eisen.

I want to ask you a direct question on this issue... So here is my question: does jDNA serve as a protective mechanism against insertional mutagenesis or not? My answer is yes. What is yours?

I honestly don't know. I'll explain why I don't know. First, I agree with Joe that you need to make your argument more quantitative.

Suppose all the functional bits of a genome are f, and all the junk is J.

Say one genome is just f, no J. Any ONE insertion might disrupt it. I get that.

Now another genome is {f,J}. Any ONE insertion is less likely to disrupt it. I get that too.

But the question is: is the TOTAL probability of insertions the same? Maybe the bigger genome you've got, the higher the probability of insertions. I don't know one way or the other.

Mathematically, let's call:

p = probability of ONE insertion in genome

q = probability that ONE insertion will be harmful

Thus the likelihood of harm is p*q.

Now my point is that as jDNA gets bigger, maybe p gets larger while q gets smaller. I don't know. It's beyond my expertise.

To be more mathematical about it, I mean that, maybe

q(genome = {f,J}) < q(genome = {f})

as you say, but then maybe

p(genome = {f,J}) > p(genome = {f})

So I don't know and can't guess what would happen to p*q. Up or down? I dunno.

Secondly, what matters, from an evolutionary point of view, is what's s, the selection coefficient, given a particular p*q? And what's the population size?

So the unknown behavior is:

What's p?
What's p*q?
What's s = selection coefficient for a given p*q?
What's the population size?

Also, Joe pointed out another problem. If a mutation occurred in a gene that produced an "accumulate junk" allele, that chromosome might accumulate a bit of junk for a few generations. But after a few generations, recombination would swap that allele to another chromosome with no junk. So the gene would lose its accumulated junk, and thus confer no advantage, immediately after recombination. It would have to start re-accumulating junk all over again.

Joe is much more of an expert on this than I, and I might not even fully understand his argument.

Joe Felsenstein said...

Thank you for the corrections. However although I agree that the more jDNA you have, the more protection you have against insertions, that is *not* the same as this being the explanation for junk DNA, because

* The selection coefficient is quite likely to be extremely small for presence of any piece of junk DNA that is not huge.

* The presence of the junk DNA (for example an active transposon) may itself increase the insertion rate.

* The insertion against which you are "protected" still occurs, but now (mostly) in the junk DNA, and it too makes more junk DNA.

So the issue is, whether the selection for junk DNA to be present explains why modest sized increases of junk DNA occur. A quantitative analysis would be needed before it would be justified to go around blogs saying that the protection against insertions is the reason why junk DNA is present.

You would need to show that the positive selection was important compared to genomic parasitism. You have not shown that. Nor have you got a way to pass the Onion Test.

Anonymous said...

Att: TheOtherJim

Thank you for your response.

In short, what you are trying to say is that mycoplasma genetic code would likely prevent the making any functional proteins in human cells. Is that what you meant?

Claudiu Bandea said...

Thanks Diogenes for your candid thoughts and insights.

You mention the need for ‘expertise’; having expertise surely helps, but often, common sense and honest evaluation of the facts are more important. Look at what happened with the ENCODE fiasco, which you covered so graphically; all that expertise involving hundreds of researchers led to a huge blunder.

Based on your comments and those of John Harshman and Joe Felsenstein, I realized that I did not explain well the two key aspects of the hypothesis: (i) the selective forces/mechanisms leading to the origin of jDNA, and (ii) the selective forces maintaining jDNA as a protective mechanism against insertional mutagenesis.

As you know, more than 40% of the human genome consists of endogenous viruses and transposable elements and much of the remaining jDNA probably contains remnants of these elements. Therefore, most of the genomic jDNA sequences originated from the spreading activity of viral elements, so there was no selection on behalf of the hosts for the origin or generation of these sequences. In other words the process or mechanism for generating jDNA sequences has been independent of selection at the host level.

However, there is a very strong host selection in regard to the regions of the genome where these jDNA sequences can insert, and also a selection on how much jDNA is maintained overall in the genome based on evolutionary constrains on increasing the genome size.

For example in bacterial and archaeal lineages, the selective pressure from inserting viral elements was so strong that it led to the evolution of specific integration sites in the genome as a defense mechanism against insertional mutagenesis. And the reason for this is that in these lineages there is very strong selective pressure against increasing the genome size, which explains the limited presence of jDNA.

In eukaryal lineages, however, the selection pressure against increasing the genome size has not been so strong, so it allowed the maintenance of certain quantities jDNA as an additional protective mechanism against insertional mutagenesis. However, even in these lineages there is a selection against ‘unlimited’ genome size; I discussed this in the comment entitled *Evolutionary constrains on genome size evolution: the Hummingbird Case* (please see: ( (Continues in the next comment)

Claudiu Bandea said...

In brief, the jDNA sequences have originated independently of the host and the selection coefficient for the maintenance of any specific jDNA sequence is very, very weak. However, the selection pressure for maintaining a certain quantity of jDNA in the genome as a defense mechanism is huge, particularly in species, such as humans, which have a long generation time. Indeed, this selection pressure takes a new dimension in organisms in which insertional mutagenesis can lead to cancer.

In humans, for example, given the enormous number of somatic cells and their high turnover rate during the reproductive span, the very larger number of insertion events, particularly by exogenous viruses such retroviruses, that would potentially lead to cancer in the absence of protective mechanisms would be evolutionarily drowning (see comment *Junk DNA and cancer: an extended and unexpected association* at: (

In conclusion, the amount of newly generated jDNA sequences is not under the host control and the selection coefficient for maintaining specific jDNA sequences is very small. However, how much of the jDNA is present at any given time in various species (including related species, such as onions) depends of the rate of jDNA generation by various mechanisms, such as spreading of viral/transposable elements, full or partial chromosomal duplications, etc., and the rate of deletion which is imposed by constrains on the overall genome size.

Obviously, there’s considerable ‘genetic drift’ in both of the processes, jDNA generation and the efficiency of deleting jDNA to a quantity that is sufficient to confer protection. This optimization process is slow, particularly, when the evolutionary constrains on genome size are relatively relaxed, which explains the huge differences in c-value for different species including onions; however, see birds particularly the hummingbirds.

Claudiu Bandea said...

Thank you for making the corrections. Please take a look at the comment I just posted above in response to Diogenes, I think it addresses some of the points you made.

Faizal Ali said...

Prof. Felsenstein:

It's like picking a scab: It can be really hard not to do it, but you know you shouldn't.

Larry's told you what you should do: Ignore.

Diogenes said...

what you are trying to say is that mycoplasma genetic code would likely prevent the making any functional proteins in human cells

This is not quite true. If a gene does not have a UGA triplet than it would function fine.

If a mycoplasma gene has a UGA triplet that should mean "Trp" is expressed with human machinery, it would truncate prematurely, which would likely snip off some domains.

If a human gene has a UGA triplet that should mean "Stop" is expressed with mycoplasma machinery, it would insert a Trp instead of stopping, and then continue transcription a bit further. Depending on the context, this might or might not kill the protein.

AllanMiller said...

If amino acid codon distributions were entirely random (they aren't, but for the sake of argument ...) a coding UGA would appear every 62nd codon on average, in a 2-STOP code. Trp -> STOP would therefore snip all proteins to an average length of 61 acids, which is almost certainly fatal.

On the other hand, STOP -> Trp would mean that proteins were extended by (on average) 31 acids, to get to one of the (again, assumed randomly distributed) remaining STOPs. This is much less likely to be universally damaging, though the chances of getting away with it genome-wide are slim.

Most variant codes vary around STOP positions, as the latter scenario - filling in a STOP - is less likely to lead to catastrophe than the reverse or an acid-for-acid swap.

TheOtherJim said...

And spelling out my caveat from the original response - this is just the codon difference. There are more differences between a human cell and a Mycoplasma's cell machinery that would probably lead to problems.

Anonymous said...

Could we agree on one thing, I hope? The mycoplasma code most likely wouldn't work in human cells like Ventor said in a video? Right?

Anonymous said...

Att: Allan Miller, TheOtherJim, Diogenes, Larry M
You guys are no doubt really good scientists, and my brother is a great scientist, and the problem I have as a very simple man, as you can tell by my posts, is how do you get over/explain/make sense of the origin of life. I can buy your evolutionary shit, because my twin brother looks like my neighbor, but the origin of life belongs to THE ONE. Please prove me wrong.

Georgi Marinov said...

The origin of life is an unsolved problem. But so were all problems that science has solved in the last few hundred years before we figured them out. Before we had the theory of evolution, the origin of everything was an unsolved problem. Not any more.

Now, admittedly, this one is a tougher nut to crack and we will likely never be able to establish (I've often joked that this problem can only be solved not by biochemists but by physicists if the latter ever invent a working time machine). But by no means does it follow from that that there was some sort of deity involved. That's the God-of-the-gaps argument in its most pathetic form

Piotr Gąsiorowski said...

It's pretty clear that Dominic is only trolling for attention and has not really attempted to understand any of the answers offered to him in good faith. He wasn't interested in the answers anyway -- he just needed some sort of justification to declare his highly original little manifesto.

AllanMiller said...

The proteins produced from mRNA would differ when processed by the two protein-coding systems, is what people have been agreeing. If you engineered human systems to add a UGA-Trp tRNA (ie, make it into the 'mycoplasma code', by that one tiny amendment), that probably would not work. If you engineered mycoplasma to remove the UGA tRNA (ie make it into the 'vertebrate code'), that almost certainly would not work.

AllanMiller said...

I can buy your evolutionary shit

Excellent! No need to fret over variant genetic codes and such then.

John Harshman said...

You can probably explain anything by a combination of ad hoc parameters, which is what you seem to be doing now. But can you devise any sort of test for your hypothesis?

SRM said...

I can buy your evolutionary shit, because my twin brother looks like my neighbor

Not sure that is evidence of evolution. Not sure what that is evidence of.

Diogenes said...


You ask a question requiring a long, technical answer.

I do not agree with Georgi that the origin of life is an unsolved problem. It's a partially solved problem.

I, like a lot of people, believe in the RNA world hypothesis. There are a lot of publications on that topic.

I feel the RNA world is so well-supported that it must be true on some level. For example: why is the ribosome a ribozyme? I can't understand the structure of the ribosome unless there was an RNA world.

There are still many huge questions about what happened before the RNA world, and how the transition from RNA to protein basis was made, and how non-free-living duplicating RNA got membranes and became free-living. But IMHO there's no doubt there was an RNA world.

If you want to believe life was created by a miracle, well, you have to incorporate your miracle around the RNA world. So you'd have to hypothesize two half-miracles, one half-miracle before the RNA world and one half-miracle to get from RNA to protein. As we figure out more steps biochemically, your half-miracles will be divided into quarter-miracles, then eighth-miracles, etc., and the power of your God will shrink with each successful experiment.

Larry disagrees with RNA world. He believes in the "Metabolism First" hypothesis. He has written on this many times.

You can google Sandwalk "Metabolism First" with quotes around the last two words, to find some of Larry's many posts on the topic.

As for the RNA world hypothesis, it's very technical and I may write on it tomorrow if I have the chance.

Anonymous said...

I don't mind hypothesis but they have to be of sound mind.
How can you believe in RNA world or metabolism first, if none of that can be tested in the lab? All I know is you can put all needed components in lab tube-RNA, mitochondria ribosome etc. all that the living cell contains to stay alive, and I guarantee you that a cell will not form out of them. No way! So what makes you so certain that this could have occurred on its own in the prebiotic soup by chance? It's totally unrealistic.

Georgi Marinov said...

One does no believe in anything, all you do is assign a probability estimate on things. RNA world seems likely because there are all sorts what seems to be relicts of the RNA world in modern cells, starting with the ribosome.

The problem with the RNA world is that the chemistry doesn't really work - the old cliche goes that the RNA world is the molecular biologist's dream and the chemist's nightmare.

Which is why I said the origin of life is an unsolved problem.

This is a good book on the subject:

Anonymous said...

"How Junk DNA Affects Heart Disease

Researchers have figured out how a mysterious DNA region previously tied to heart disease may exert its effect. The discovery could open the door to new prevention and treatment strategies.

Heart disease is the leading cause of death in the United States and a major cause of disability. Genome-wide association studies, which scan the genomes of large numbers of people to find genetic variations associated with disease, recently linked gene variations on human chromosome 9p21 with an increased risk for heart disease. The variants only moderately raise the risk of disease, but they’re very common and so may make a significant contribution to heart disease in the population.

How these variants affect heart disease has been a mystery. They aren't associated with known risk factors, such as blood lipid levels, hypertension or diabetes. They're also within a 58,000-base stretch of so-called "junk" DNA, which contains no known protein-coding genes. The human genome has about 3 billion human DNA base pairs, with genes making up only a small fraction of that—about 2%. Little is known about the function of the remaining 98% that doesn't code for proteins.

Scientists from the U.S. Department of Energy’s Lawrence Berkeley National Laboratory led by Dr. Len A. Pennacchio set out to try to understand how this non-coding interval between genes affects heart disease. To do this, they genetically engineered mice to completely lack the interval. The study, partly funded by NIH's National Heart, Lung and Blood Institute (NHLBI) and National Human Genome Research Institute (NHGRI), appeared in the advance online edition of Nature on February 21, 2010.

The researchers looked at genes near the deleted interval and found that 2 genes—Cdkn2a and Cdkn2b—were expressed at 10-fold lower levels in heart tissue of altered mice. Cdkn2a and Cdkn2b belong to a family of "cell cycle" genes, which are involved in regulating cell proliferation. When the researchers examined how muscle cells taken from the heart grew in the laboratory, they found that cells lacking the interval multiplied faster than controls and kept proliferating after the control cells had stopped.

When the scientists placed 40 of the genetically altered mice and 40 controls on a high-fat, high-cholesterol diet for 20 weeks, blood lipid levels rose to similar levels in both groups of mice. There were also no significant differences in fatty deposits in arteries. However, the diet caused substantially more deaths among the mice lacking the non-coding interval. These results suggest that the interval exerts its effect through a mechanism independent of blood lipid levels and other known risk factors.

"We show that this non-coding interval affects the expression of 2 cell cycle inhibitor genes located almost 100,000 base pairs away from the deletion," Pennacchio says. "We believe that something goes awry in variants of this interval, causing vascular cells to divide and multiply more quickly than usual." That may increase the risk for heart disease, the researchers speculate, by narrowing coronary arteries.

"Non-coding DNA is a huge area of the genome, waiting to be explored, which could have huge dividends for understanding and treating disease," Pennacchio adds.

—by Harrison Wein, Ph.D."

Georgi Marinov said...

1. Enhancers are not junk DNA and never have been

2. Apparently you have neither followed this blog for long nor are you in general familiar with the mismatch between the hyperbolism of press releases and the reality of the research they report on

3. The reason that one paper was exciting is that few people have bothered to knock out enhancers - until very recently, it was quite a difficult technical feat to do just one of them and since there are reasons to think the effects of such a knock will often not be very great, as genes are usually controlled by multiple enhancers, sometimes active in different cells but also often redundant with each other, the incentive to do this has not been great. Pennacchio has been one of the very few to do that. This will change as now we have much more powerful tools for knocking out things - first TALENs, and more recently, and especially relevant if one wants to do this high-throughput, the CRISPR-based protocols for direct genome editing.

So you will see a lot more of that type of work in the future. But this will not change the first point I made - enhancers are not junk DNA and have never been considered such, which is why such press releases are deeply misleading.

Georgi Marinov said...

P.S. Enhancers also tend to be conserved. Not always, there is much higher turnover of enhancer elements than of coding sequences, but nevertheless, one can find a lot of them through comparative genomics alone. And people have been doing that for a decade, ever since whole genome sequences became available.

Anonymous said...

This is from Nature:

"Knowles cautions that the study doesn't prove that non-coding DNA has no function. "Those mice were alive, that's what we know about them," she says. "We don't know if they have abnormalities that we don't test for."

David Haussler of the University of California, Santa Cruz, who has investigated why genetic regions are conserved, says that Rubin's study gives no hint that the deleted DNA has a function. But he also believes that non-coding regions may have an effect too subtle to be picked up in the tests to far.

"Survival in the laboratory for a generation or two is not the same as successful competition in the wild for millions of years," he argues. "Darwinian selection is a tougher test."

Link to the article:

Diogenes said...


How can you believe in RNA world or metabolism first, if none of that can be tested in the lab?

Who said it can't be tested in the lab? Some creationist told you that, and you believed them.

However, your idea for how to test it is absurd, and I think deliberately absurd.

... you can put all needed components in lab tube-RNA, mitochondria ribosome etc. all that the living cell contains to stay alive, and I guarantee you that a cell will not form out of them.

Riiight. You got us. Atoms in a blender does not make a living cell.

Everybody agrees that multiple steps would be involved, and a free-living protein-based cell comes at the end of all those steps.

The point of laboratory tests is not to make a living cell by randomly throwing together whatever, but to investigate particular steps in the transition from abiotic chemistry to the RNA world, and then from RNA world to membrane synthesis. Where metabolism comes in, before or after RNA, is disputed.

Your idea that the ingredients in step 1 mixed together in a blender will make the results of step 30 is ridiculous, and it's deliberately ridiculous-- you chose it to misrepresent how science is done.

If you want to trash-talk, OK, we'll win at trash-talking. Since you believe in the Bible, you think dirt turned into the people. Put some dirt in a blender and see if it turns into a naked man and woman.

One wonders what point is there giving technical answers to creationists? They don't want to learn science, they want to stick science in a blender and hit puree.

Georgi Marinov said...

you can put all needed components in lab tube-RNA, mitochondria ribosome etc.

I didn't read that comment before - did he really said "mitochondria"???????? Sigh...

Diogenes said...

This is from the idiotic article cited by Dominic.

They're also within a 58,000-base stretch of so-called "junk" DNA, which contains no known protein-coding genes.

For the ten millionth time, NO ONE EVER SAID NON-CODING DNA = JUNK.

The fact that the author thinks geneticists believed non-coding DNA = Junk indicates he is an IDIOT. Perhaps not an IDiot, but certainly an IDIOT.

How many times do I have to copy the list of Nobel Prizes handed out in the last 50 years to scientists who found novel functions in non-coding DNA?

Georgi: Enhancers are not junk DNA and never have been

Thank you Georgi!

Mikkel Rumraket Rasmussen said...

This is what happens when someone who's essentially completely clueless starts babbling on the subject.

And nobody believes in "prebiotic soup". Creationism: still stuck in the 1800's.

Whether RNA came "first", directly from abiotic geochemistry, or it didn't, is immaterial with respect to the fact that we're pretty sure there was an RNA world from which DNA later evolved. What really came before this RNA world is what we don't know, and some have suggested RNA was the very first thing itself, others have suggested it was preceded by metabolism or even simpler kinds of genetic polymers.
This is what's wonderful about science, because instead of just giving up and declaring goddidit, scientists are trying to find out how it happened.

Mikkel Rumraket Rasmussen said...

Also.. what is it with creationism and the origin of life always being introduced in discussions on evolution? This topic is about Junk DNA, and there's a creationist babbling about the origin of life?

Mikkel Rumraket Rasmussen said...

I can buy your evolutionary shit, because my twin brother looks like my neighbor, but the origin of life belongs to THE ONE. Please prove me wrong.

Theism, attempting to shift the burden of proof since 5000 BCE.

Here's how it works dominic. YOu make a claim, you support it. Please demonstrate instantaneous magical creation of a living organism.

SRM said...

Please demonstrate instantaneous magical creation of a living organism.

Dominic will need to acquire a GOD blender, before he hits puree.

Claudiu Bandea said...


In the original paper ( and more recent comments (, I proposed an experimental approach that would address a highly significant medical aspect of the hypothesis; jDNA protection against cancer. You might want to evaluate it.

It would help to support your assertion that I used “ad hoc parameters” by specifically pointing them out.

Also, it seems that you don’t’ agree that in humans, for example, jDNA protects against inspectional mutagenesis. Does jDNA protects against inspectional mutagenesis, or not?

Claudiu Bandea said...

Joe Felsenstein: “ Thank you for the corrections

It seems that this fellow lutesuite, who has a hidden identity, did not understand you message. He based his derogatory comment on your earliest incorrect statement (see above), so you might want to ask him to apologize.

Also, I don’t think that you take directions from him or Lary on your comments, so I think he should apologize for that, too!

Anonymous said...

A funny thing about the origin of life is that some believe that life originated spontaneously, by accident etc. Then, it remained simple for millions or billions of years but as soon as it got more complex and became a multicellular organism, life decided to invent death. What a damn luck! Had life not been so inventive, we could live forever!

Anonymous said...


Sure, "spontaneously" must be much more far fetched than by magic performed by a being that only exists in people's imagination, and whose complexity, desires, and magic powers don't require any rational explanation for their origins.

What makes you think that there was no death before multicellularity? Do you truly believe that the first cell(s) is(are) still floating around?

Mikkel Rumraket Rasmussen said...

@Negative Entropy
Dominic doesn't know what the words spontaneous and by accident means. To him, he seriously thinks there are people who imagine a living cell miraculously assembling itself from scratch, I see this with creationists all the time. I don't know how they got this idiotic idea to begin with, but they all regurgitate it ignorantly repeatedly as if by training.
Instead of dealing with the actual science, they're bent on the least charitable caricatures and strawmen they can possibly conjure up. And here I thought that would be in violation of the 9th commandment, silly me.

"Life decided to invent death"
Death is not an invention, it's an inevitability. When you run out of nutrients, you die and decompose. Sooner or later, every organism is going to run into either food scarcity, disease, natural/environmental disaster or a predator. It's that simple.

nmanning said...

A bit tangential, but I really love the date of Dembski's "prediction" on function in junk DNA...

Anonymous said...

Sorry guys but I was not only talking about the invention of death but also about the programmed cell death that seems to me to be just a work of art; It is so fine tuned you scientist have no idea about that. I'm sure you can come up with some testable shitheories, but so can I.

Robin said...

Hmm... so the official figure now is that 90% of our DNA is junk. I used to hear that 98% was junk and then in 2009 95% was junk. Now is it is 90% and I have heard the figure 80% being bandied around (for example Dr Merlin Crossley on the Australian ABC's "Ockhams Razor"

I suppose that is the beauty of science, facts are only as good as the latest experiment.

But if this keeps up then it will take a lot less than 9,000 years to find a function for all DNA.

Georgi Marinov said...

Where are you getting these figures from???

The hard numbers are the following, which I just pulled out of the latest GENCODE version (v16) and from UCSC's repeatMasker:

2.89% exons of protein coding genes (this includes UTRs, which contribute a significant fraction of this percentage)
0.20% lncRNAs
0.0087% miRNAs
~0.003% structural RNAs (rRNA, tRNA, snoRNA, snRNA, etc)
0.20% other ncRNAs

In total, 3.33% exons

In addition to that, there are 0.5% pseudogenes of various types

Then we have:

3.15% DNA repeats
20.28% LINE repeats
8.52% LTR repeats
12.6% SINE repeats

plus a tiny percentage of other types of repeats (defective copies of structural RNAs, etc.)

A total of 46.45% annotated repetitive elements

In addition to this, there is a single-digit percentage of conserved non-exonic, non-repetitive elements, the exact figure of which varies depending on how you define conserved. Which is mostly regulatory elements, but as a whole it has not been annotated outside of what ENCODE has done, and ENCODE has not really annotated those yet, it has just generated the maps that will provide the foundation for that annotation in the coming years.

The earliest GENCODE version I have is v3c, which is from around 2009, in that one there were 2.5% exons of protein coding genes, and the 0.2% of lncRNAs were not annotated. That's a growth of 0.5%, a significant portion of which is of dubious significance

So where are you getting the idea that it was 95% functional DNA in 2009, 90% now and soon to become 80%?

And do you seriously think those 46% of annotated repetitive elements (people like to talk about how there is a lot of functional DNA left to annotate, but there is likely much more DNA consisting of repeat copies so broken down and degenerated that the repeat annotation programs simply have not picked them up) will somehow magically turn out to be vitally important functional elements??

Reuben Alvesson said...

Hi, an student of mine who is also a programming genius asked me this question after reading details of the Encode project:

If all the DNA is functional and it contains cascading step-by-step instructions meaning that each step depends of the results of the preceding one to continue the chain of instructions for assembling the embryos of animals then how the Evolution mechanism of changing one species into another (individual random non-guided mutations to the DNA) would be able to change this chain instructions without ruining it ? He told me that if you change one letter in a computer code it is ruined and does not work since all instructions depend on the precide execution of the preceding ones....Thanks for your help

Larry Moran said...

Your programing genius student is correct. If biological organisms were as well-designed as perfect computer code such that changing even one letter would break the code, then evolution would be impossible.

Topgoosz said...

So, what will be the result when you throw the well designed computer in the water?
Thats why God create live in a way it can survive"under any curcimstances within the edges God ordered. That's what i call well-designed. God gives live a downgrade caused by sin. Junk-DNA fits in my opinion perfect in the case of creation.

Why do you think you get older, sick weak and eventually die? Junk DNA i
s a misinterpretation. In reality you have to call it ''the results of a downgrade''.