More Recent Comments

Monday, March 16, 2009

Casey Luskin on Junk DNA and Junk RNA

 
Intelligent Design Creationists can't abide junk DNA. Its very existence refutes the idea that living things are designed by some intelligent being. This is why the IDiots go out of their way to make up stories "disproving" junk DNA.

The latest attempt is by Casey Luskin [Nature Paper Shows "Junk-RNA" Going the Same Direction as "Junk-DNA"]. Having failed to explain why half of the human genome is composed of defective transposons, he now pins his hope on the idea that most of the genome is transcribed. Luskin seems particularly upset by my statement that most of these transcripts are junk [Junk RNA].

Luskin thinks that a recent paper in Nature supports his view that a large fraction of the genome isn't junk. The paper by Guttman et al. (2009) says no such thing. Here's the important part ...
Genomic projects over the past decade have used shotgun sequencing and microarray hybridization1, 2, 3, 4 to obtain evidence for many thousands of additional non-coding transcripts in mammals. Although the number of transcripts has grown, so too have the doubts as to whether most are biologically functional5, 6, 13. The main concern was raised by the observation that most of the intergenic transcripts show little to no evolutionary conservation5, 13. Strictly speaking, the absence of evolutionary conservation cannot prove the absence of function. But, the markedly low rate of conservation seen in the current catalogues of large non-coding transcripts (<5% of cases) is unprecedented and would require that each mammalian clade evolves its own distinct repertoire of non-coding transcripts. Instead, the data suggest that the current catalogues may consist largely of transcriptional noise, with a minority of bona fide functional lincRNAs hidden amid this background. Thus, to expand our understanding of functional lincRNAs, we are faced with two important challenges: (1) identifying lincRNAs that are most likely to be functional; and (2) inferring putative functions for these lincRNAs that can be tested in hypothesis-driven experiments.
In other words, most of the transcripts are probably transcriptional noise, or junk, just as I said. This is the consensus opinion among informed1 molecular biologists.

Guttman et al. wanted to identify the small subset that might be functional. They identified 1,675 transcripts that show evidence of conservation. The average transcript has six exons averaging 250 bp. Thus, each transcript has about 1500 bp. of conserved exon sequence.

Even if every single one of these lincRNAs have a biological function they will only account for 1675 × 1500 = 2.5 million bp. This represents less than 0.1% of the genome. Casey Luskin ain't gonna disprove junk DNA using this paper.

Luskin ends his article with ...
As an ID proponent, I'm still waiting for Darwinists to let go of their precious "junk" arguments for blind evolution and common descent and learn the lesson that you can't assume that if we don't yet see function for a biomolecule, then it's probably just "junk."
This is a point of view that creationists share with many scientists who haven't studied the subject. They assume that the only reason for labeling most of our DNA junk is because we don't know what it does. That's just not true. There's plenty of good evidence that most of our genome can't be functional. We know a lot about the part that consists of transposons and defective transposons, for example [Junk in Your Genome: SINES and Junk in your Genome: LINEs]. That's 44% of our genome.


1. I added the qualifier "informed" after a commenter pointed out that most molecular biologists probably don't know enough about the topic to have an opinion. Thus, according to this commenter, the consensus opinion would be "I don't know."

39 comments :

Anonymous said...

This is the consensus opinion among molecular biologists.

No, this is your opinion. If you have actual data or a citation quantifying the number of molecular biologists that do or do not hold this opinion, please share the data or cite the reference. I'm a molecular biologist, and I'm agnostic on the junkness of junk, so that makes N=2 anyway.

Anonymous said...

Irony Meter Explosion Alert.

Luskin is criticizing molecular biologists, saying that just because you haven't found a function for a unit of genetic material, it doesn't meant it's functionless junk. Why is this line of thinking always out taking a wiz whenever Luskin is claiming that if science hasn't yet figured something out, God must've done it.

Alex said...

I'm sorry, but this is a ridiculous standard of evidence:
If you have actual data or a citation quantifying the number of molecular biologists that do or do not hold this opinion, please share the data or cite the reference.

That is all.

NickM said...

Heh. The paper specifically says that 95% of the transcribed sequences aren't conserved. This, along with other evidence, led to the suggestion that much of the transcribed sequence was noise. The paper then looks at *some* of the stuff from the 5% that are somewhat conserved, and finds function. Luskin concludes: everythin, conserved or not, has function and Larry is clearly wrong!

ERV said...

Anon-- If you are a 'molecular biologist', why are you acting like a twat? Youve got about 34K recognizably retroviral genes. A handful of them have the gross capability of coding for a functional protein, and then we have, like, two that have been domesticated for non-retroviral functions.

How, exactly, are those ~34K retroviral genes not junk? I mean, theyre handy points for chromosomal rearrangements/evolution, ergo genomic instability, ergo cancer, but thats hardly the kind of use Casey TARD is going for.

How the hell can you be an 'agnostic' on this issue, as a 'molecular biologist', anon?

Seth said...

Anon-
1- Casey Luskin is an attorney, not a biologist, so N=1
2- "agnostic" is not a synonym for uninformed.

If you are a molecular biologist, use your knowledge to give context to the available literature, and form an opinion.

Anonymous said...

I wasn't counting Luskin, whoever he is, just me and Larry.

I'd wager a cheeseburger that this isn't even the consensus opinion in Larry's department.

Show of hands next faculty meeting?

True, false, or don't know: Over 90% of the human genome has no useful function whatsoever.

My cheeseburger's on "don't know".

ERV said...

After reading this paper Im calling out 'anon' as a lying little shit.

You are not a molecular biologist.

If you were, you would have easy access to read this paper yourself, including their references to 'Larrys' consensus opinion, which they address explicitly in the abstract, intro, and discussion.

LOL! Creationist troll needs better camouflage. Fail.

Billy said...

Creationist troll!

Why are there telomeric sequences inside our 2nd chromosome?

Seth said...

True, false, or don't know: Some of the human genome has no useful function whatsoever.

Fixed it. Answer will be a resounding yes.

Why are there telomeric sequences in human Chr2? Because, like the cheeseburger I just won from Anon, god made it a double stacker. I bet we find lettuce and tomato in the centromere.

John Farrell said...

Excellent post! Taken together, Larry, I think you can (and should) make a nice little book out of all your posts on junk DNA.

Billy said...

I bet we find lettuce and tomato in the centromere

Nah, we find a nice ripe tomato and a squishy rotten one where the other centromere was (there's a clue there for troll boy)

Larry Moran said...

anonymous says,

I'm a molecular biologist, and I'm agnostic on the junkness of junk

Point taken. Most molecular biologists don't know enough about controversial topics to have an informed opinion.

I agree with you that the consensus among the entire group of molecular biologists would have to be "I don't know."

I changed the posting to "informed molecular biologists." By that I meant those who have at least read and understood Kevin Struhl's paper Transcriptional noise and the fidelity of initiation by RNA polymerase II.

Anonymous said...

I've got a candidate for useless DNA so obvious and simple even my fellow attorney Casey Luskin should understand it: The part of the genome responsible for making nipples on guys.

Yes, of course it is ridiculous to consider this "junk" in the context of what we know about development, but what if, like Casey, you figure all our genes are the product of a supersmart, even divine, designer?

Anonymous said...

I'm a molecular biologist, not a troll, ERV on the other hand......

Here's something for those who don't buy into Larry's junk crap:

http://biopinionated.com/2008/10/28/hammering-nails-in-the-junk-dna-coffin/

N = N + 1

Billy said...

ome interesting references there Nils. None of which prove that all DNA has a function. The interesting thing is that some sequences apopear to have gained function - such as the evolution of a couple of P53 binding sites from Alus.
With a new LINE per 100 births, are you claiming that these are somehow functional and necessary?

ERV said...

"I'm a molecular biologist..."
No, youre not. I was writing about the evolutionary capabilities of Alus and ERV domestication before I even started grad school. The fact, like, 2 ERV genes have been domesticated doesnt negate the fact 34,000 are leftover garbage. And did you even read the papers cited in your link? Im guessing no, because I have read many of them, and they do not say what you think they say-- Much like the paper Casey TARD wrote about.

While Ive always been precocious, I am not 'smarter' or 'better read' than professionals in my field.

So why did I know this stuff years ago and youre a 'molecular biologist' presenting it like a divine revelation?

You either A) are not a molecular biologist, or B) need to reacquaint yourself with PubMed.

Lurk more.

Billy said...

You either A) are not a molecular biologist, or B) need to reacquaint yourself with PubMed.


Or C he graduated from somewhere like Liberty "university". I've come across some of their "molecular biologists" before - quite sad! I recall that one of their main faculty members has something like 6 papers in minor journals - no doubt a coup by their standards.

I do doubt he read the papers too.

Anonymous said...

Put Struhl down in the "don't know" category too, although clearly he has an opinion and a testable hypothesis.
However, the relative proportions of biologically significant noncoding RNAs and transcriptional noise are unknown. etc.

He's calling for more work to be done, which would be unnecessary if the consensus opinion among "informed" molecular biologists was that the question was already answered. In fact, Struhl's entire opinion paper would be redundant if the field was really as heavily convinced as some seem to believe.

To throw in a little molecular biology for the doubters, a specificity of 10e4 between an optimal transcriptional start site and a random genomic transcriptional noise site is very unimpressive for a multi-component protein machine, hardly more specific than a generic restriction enzyme. Yeast is clearly capable of doing much much better, even at the level of single molecules, for example the HO endonuclease responsible for mating-type switching has one recognition site in the 1.2e7 bp yeast genome.

Struhl's argument also hinges on the idea that chromatin associated RNAP phosphorylated on the CTD is in fact elongating, but that RNA thereby produced is not detectable because it is rapidly degraded by nonsense-mediated decay, or other mechanisms. While plausible, the majority of yeast genes are unspliced and therefore not subject to nonsense-mediated decay, which complicates this assumption. Are the non-specific transcripts in general spliced? Further study is clearly warranted.

Anonymous said...

ERV and Billy. Your arguments are mostly just sad (Resorting to name-calling and university bashing will probably not benefit your cause). Billy: I do not think anyone is saying that all DNA has a function, but what many are saying is that a lot of what Larry claims to be junk probably isn't. ERV: not a revelation of any sort, just balancing out the "90 % of our genome is junk" crap.

Billy said...

So Nils - Liberty graduate?

Now, perhaps you can tell me what is sad about pointing out that your references hardly support your case that most LINES, SINES etc are not junk. Infact, your "rgument" involves calling a position "crap" without the evidence to back that up. You have a handfull of papers- I wonder what proportion of the junk genome that covers. I am open minded about how much is actually junk, but a lot of it looks like junk.......

Anonymous said...

Liberty Professor no less, thank you very much (costs a bit more, but sure is worth the money).

I have argued on the proportion in the posts leading up to the post you commented on previously, go to http://biopinionated.com/tag/junk-dna/ for more.

My main point is exactly your point, but in reverse. The 90 % of the genome argument is based on a lack of function (or evidence of function if you will). I am arguing that you cannot conclude with numbers based on that, but this is precisely what Larry has done. Presently, nobody knows exactly how much of our genome that constitutes junk. A true scientist would say "OK, let's try and see if we can find some function then" (which is an approach that will bring us more knowledge), and not "Oh, it's all junk, forget about it" (which leads us nowhere).

Billy said...

Liberty Professor no less, thank you very much (costs a bit more, but sure is worth the money).


I saw an advert for a biologist there once - you had to be a YEC - not really encouraging critical thought is it?

Anyway, let's see what you've got.

The 90 % of the genome argument is based on a lack of function (or evidence of function if you will).

Well, we know enough about silencing mutations etc to make a reasonable case as to whether something may or may not have a function. The problem with your approach is that (and i'm sure you will correct me if I am wrong) is that you seem to be assuming something should have a function. I would say you have to keep an open mind one way or another. Is it also not reasonable to assume that the intrachromosomal telomeres of chromosome 2 should have no function? What about the GLUO pseudogene? There definately is evidence of something being non functional. What about cetacean olfactory receptors or notothenoid haemoglobin genes? It is not just an assumption that something does not have a function - there are good reasons to think that - eg mutations, lack of transcript and chromosomal position when it comes to "fossil" genes.
Personally, I will evaluate my position on how much junk there is based on evidence as it comes in. Are you willing to do the same, or do you have YEC beliefs that prevent you from doing so?

Is it your position that junk actually exists?

Anonymous said...

Billy.

1) oops, did you take the Liberty thing seriously ? I've never heard of the place and....What's a YEC ? :-)

2) YES "I will evaluate my position on how much junk there is based on evidence as it comes in." That has been my point all along.

Doppelganger said...

Nils,
That you cite Andras Pellionisz' historical revisionism-littered disaster of a paper in your litany of citations purportedly showing that there is less 'junk-DNA' than believed tells me that you don't pay enough attention to the content of the papers on your list.

Anonymous said...

Doppelganger. I know Pellionisz is controversial (to put it mildly). I have posted that any comments to the list is welcome, - yours is noted.

Billy said...

1) oops, did you take the Liberty thing seriously ? I've never heard of the place and....What's a YEC ? :-)


Not after I checked your site.

YEC : Young Earth Creationist

Here is the Liberty ad:

Position: Multiple Faculty Positions in the 2007-2008
Salary: Unspecified
Institution: Liberty University
Location: Virginia
Date posted: 10/30/2006

Liberty University

Liberty University is a Christian liberal arts university with nearly 9,600 resident students as well as several thousand additional students in its distance-learning program. The University is seeking faculty who can demonstrate a personal faith commitment to its evangelical Christian purpose and who are dedicated to excellence in teaching for the following anticipated openings in the 2007-2008 academic year.

COLLEGE OF ARTS AND SCIENCES - Dr. Roger Schultz, Dean

Biology: Two positions. Rank open. Ph.D. and compatibility with a young-earth creationist philosophy required. 1) Human anatomy and physiology, 2) undergraduate genetics. Supervision of undergraduate research expected. Contact Dr. Paul Sattler - pwsattler@liberty.edu.

Center for Creation Studies: Rank open. Ph.D. and experience in the origins controversy from a young earth creation perspective required.

Faculty will teach the required course in creation as well as develop and/or teach in their area of science expertise. Ability to teach courses in astronomy, anatomy and physiology, or other biology courses preferred. Contact Dr. David DeWitt dadewitt@liberty.edu.


I love the way they obviously take their astronomy teaching seriously. No doubt it is good od biblical geo-centric astronomy

Doppelganger said...

I'm not sure that 'controversial' really engenders the true feeling most competwent scientsts have when they read his paper.

It is childish, the figures contain typos, it is petty, it contains flat out lies, and the main theme has ZERO support. Why it is publication is a mystery, how it got past review smacks of a Sternberg-Meyer event.

If you want to leave it on your list, feel free. But realize that doing so calls your scientific judgement into question.

Anonymous said...

*rolls eyes*
Compare the genome sizes of Fugu, Homo, and Ambystoma and then tell me there's no junk DNA. Jeez!

Anonymous said...

If you have actual data or a citation quantifying the number of molecular biologists that do or do not hold this opinion...

Science is not a democracy. It doesn't matter how many people think something is or isn't true for things other than cultural exchanges. Facts and facts and that's the end of it. If bits of DNA are biochemically incapable of coding anything, it means that they have no real function and no amount of fervent belief otherwise will suddenly make them coding genes.

Anonymous said...

I'm not a molecular biologist, so maybe this question is naive, but how exactly are we defining "no function" anyways?

Would stuff like "makes DNA strand longer so distant regulatory element can more easily loop around and be placed near target gene during transcription" or "provides a 'decoy' target for retroviral insertions to reduce the risk of retroviral infection disrupting function of important gene or causing cancer" or "produces unstable region of chromosome increasing probability of transposition events with the potential to generate speciation, thereby increasing the evolution of evolvability", etc count as "function"?

If you're willing to define "function" loosely enough, and disregard any consideration of optimization, then almost everything can be said to have "some" function, however minuscule.

Freidenker85 said...

Guys, guys. No one in the comment section claimed for a minute that junk DNA doesn't exist. The only thing that's even contentious is the AMOUNT of junk that exists - but as far as this post is concerned, this doesn't matter. It takes an iota of junk to make God (come on, we all know who the "designer" is) appear like a twerp and have the entire ID position go down the toilet.

No one here is denying that junk exists but some feeble-hearted twerp from the DI, so it's time everyone in the comment section buys everyone else a beer.

Anonymous said...

The existence of junk DNA doesn't by itself necessarily disprove ID, although it does demonstrate that the putative designer would have to be remarkably inefficient and lacking in foresight. (Certainly not omnipotent, omniscient, or benevolent, and not even particularly intelligent)

But the specific sequences in the junk DNA and their distribution throughout the genome of each individual lineage is a clear indication of common descent. It also demonstrates unequivocably the processes that increase information content in the genome, and shows that these processes are random in nature.

So it would also suggest that the putative designer is a liar who deliberately made things to appear as if they evolved.

Freidenker85 said...
This comment has been removed by the author.
Freidenker85 said...

Oh, so "intelligent designer" can also mean "rather intelligent, but not perfect, designer"?

Okay. I don't know what's worse for the ID crowd, that God doesn't exist or that God's stupid.

Larry Moran said...

amphiox asks,

Would stuff like "makes DNA strand longer so distant regulatory element can more easily loop around and be placed near target gene during transcription"

That would count as a function. We know this isn't the case because most loops are only a few hundred bp. Bacteria and yeast have the same mechanism of transcription regulation and they get along just fine without large amounts of junk DNA.

or "provides a 'decoy' target for retroviral insertions to reduce the risk of retroviral infection disrupting function of important gene or causing cancer"

This doesn't make any sense. Once the large "decoy" regions become saturated with retroviruses and transposons, which they are, they simply serve as a reservoir for more transposons and retroviruses. This makes the problem worse, not better.

or "produces unstable region of chromosome increasing probability of transposition events with the potential to generate speciation, thereby increasing the evolution of evolvability", etc count as "function"?

Evolution cannot plan for the future. You can't have selection for something that might be beneficial a million yeas form now.

If you're willing to define "function" loosely enough, and disregard any consideration of optimization, then almost everything can be said to have "some" function, however minuscule.

That's correct. But so far nobody has discovered any such function for most of our DNA.

Anonymous said...

Evolution cannot plan for the future. You can't have selection for something that might be beneficial a million years from now.

Look at it this way: Evolution is the result of iterative cycles of generation of phenotypic diversity followed by reduction of phenotypic diversity.

Phenotypic differences are usually the result of genotypic differences. The generation of genotypic diversity is usually the result of random genetic alteration. The reduction in phenotypic diversity is stochastic but influenced by differences in relative fitness, either internal or external in nature.

When fitness differences do not exist, the reduction in phenotypic diversity is called "drift". When fitness differences do exist, the reduction in phenotypic diversity is called "selection".

When humans observe the generation of specific phenotypic subsets of a phenotypically diverse population that later persist following externally mediated reductions in phenotypic diversity, humans tend to anthropomorphize this observation in terms of past planning for future events, leading to the erroneous inference that a "planner" exists, when in reality, having a sufficiently large pre-existing phenotypic diversity in the population is what allowed seemingly "intelligent" response to new selective forces or better response to existing forces (see the Luria-Delbruck experiment for details).

"You can't have selection for something that might be beneficial a million years from now," but you can have future selection for something that by chance might be generated today. For this reason, it is advantageous for species to maintain a reasonable degree of genomic diversity to be pre-enabled to respond to unpredictable environmental forces. This means that mechanisms that enable generation of genomic diversity can potentially confer an evolutionary advantage.

What then is the evolutionary significance (if any) of "junk DNA and RNA"? It seems unlikely that the junk confers a direct increase in fitness. Does the junk facilitate the generation of genomic or phenotypic diversity? This seems less obviously true or false.

Anonymous said...

means that mechanisms that enable generation of genomic diversity can potentially confer an evolutionary advantage.

Offset against the risk of just causing cancer -- the normal result of chromosomal abnormalities.

Anonymous said...

Offset against the risk of just causing cancer -- the normal result of chromosomal abnormalities.

Cancer isn't an issue for single celled organisms. For the multicellular organisms, it is necessary to distinguish between meiotic chromosomal alterations and mitotic chromosomal aberrations. In general, the meiotic ones are relevant for evolution, the mitotic ones are relevant for cancer. The normal result of large-scale meiotic chromosomal abnormalities is embryonic lethality, not cancer.