More Recent Comments

Friday, January 04, 2013

Science Magazine Chooses ENCODE Results as One of the Top Ten Breakthroughs in 2012

Science magazine (published by AAAS) was one of the major news sources that fell hook, line and sinker for the ENCODE/Nature publicity campaign last September [Science Writes Eulogy for Junk DNA]. It even published a laudatory three page profile of Ewan Birney, the man responsible for misrepresenting the ENCODE results as evidence that most of our genome is functional [Ewan Birney: Genomics' Big Talker].

I was somewhat apprehensive when I saw that the editors of Science had picked the ENCODE results as one of the top ten breakthroughs [Genomics Beyond Genes]. Would the editors continue to promote the idea that most of the human genome is functional?

The short answer is "yes," as you can see from the summary box below. The person who wrote the article says that "the functional portion of the genome" is 80% of the total. He/she emphasizes this in the text ...
ENCODE researchers took an intensive look not just at genes but at all of the DNA in between. Their results drive home that much of the genome that at one time was dismissed as "junk DNA" actually seems to play an essential role, often by helping to turn genes on or off. They pinpointed hundreds of thousands of landing spots for proteins that influence gene activity, many thousands of stretches of DNA that code for different types of RNA, and lots of places where chemical modifications serve to silence stretches of our chromosomes, concluding that 80% of the genome was biochemically active.
This is, as we now know, very misleading. Just because DNA binding proteins are found on DNA does not mean that the binding site has a function. Just because a region of DNA is transcribed on occasion does not mean that the RNA is functional. You do not overthrow 40 years of evidence that most of our genome is junk on the basis of spurious binding sites and aberrant transcription.

Besides, we already knew about all these results years ago so the September papers do not count as a real "breakthrough" by any stretch of the imagination. But Science justifies the "breakthrough" status by pointing out that major newspapers were fooled ...
When these papers were published in September, the media went wild. ENCODE was hailed in The New York Times as a "stunning resource" and "a major medical and scientific breakthrough" with enormous and immediate implications for human health. The Guardian called it "the most significant shift in scientists' understanding of the way our DNA operates since the sequencing of the human genome."
This is a cop-out. Science is published by a scientific organization and we expect a more serious appraisal than what you might see in newspapers. What do the editors of Science think? Do they also believe that the ENCODE results represent a "significant shift in scientists' understanding of the way our DNA operates"? 'Cause if they do, they are as wrong as the journalists who wrote for The Guardian.

In fairness, someone at Science decided that it would be a good idea to mention the controversy, so they added
But several scientists in the blogosphere called the coverage overhyped and blamed the journals and ENCODE leaders for overplaying the significance of the results. For example, ENCODE reported that 76% of DNA is transcribed to RNA, most of which does not go on to help make proteins. Various RNAs home in on different cell compartments, as if they have fixed addresses where they operate, suggesting that they play a role in the cell. Critics argue, however, that it was already known that a lot of RNA was made, and that many of these RNAs may be spurious genome products that serve no purpose. Likewise, one ENCODE researcher found 3.9 million regions across 349 types of cells where proteins called transcription factors bind to the genome—but again, it's unclear how much of that binding is functional.
I suppose we have to give them credit for recognizing that there's a problem but it's clear that the person who wrote the article doesn't really believe the critics otherwise he/she would never have led off with this statement ...
Their results drive home that much of the genome that at one time was dismissed as "junk DNA" actually seems to play an essential role, often by helping to turn genes on or off.
All in all, this is just another example of bad science journalism. Science magazine is shirking its responsibility as a leading science magazine. Either the ENCODE publicists are correct or the scientists in the blogosphere are correct. They can't both be right. The editors of Science owe it to their readers to get to the bottom of this controversy.

Is it, or is it not, true that most of the human genome is functional? It still looks to me like the editors have chosen the wrong side of the controversy.


20 comments :

Blas said...

But that is peer reviewed publications, science at his top. isn`t it?
Are you testing your own medicine?

invivoMark said...

I can't tell if it's just because I'm a very new and naive scientist, but I feel like I'm quickly losing trust in Science and Nature - like they've become so obsessed with publishing the big stories that they've tossed basic scientific rigor out the window. Surely, if this is how they are now, they haven't always been so bad?

And I wish I was just talking about the few major media-frenzy items, like ENCODE and arsenic bacteria. But I've seen a few other rotten apples coming out that didn't get the same press coverage, but were almost as bad at doing what they set out to do.

I'm not even sure I'd want to publish in Science or Nature any more.

Diogenes said...

Is it, or is it not, true that most of the human genome is functional?

No. The question is what is the definition of "function" that is relevant to the Junk DNA hypothesis.

Transcription to RNA is not "function" by any definition relevant to the Junk DNA hypothesis.

Chemical modification and binding proteins MAY be relevant the Junk DNA hypothesis, but in an reverse sense from that meant by Junk DNA deniers. That is, chemical modification and binding proteins may either be random noise applied to random sequence matches, or they may have the "function" of silencing transcription of old transposons or viral DNA. That is also a "function", but not "function" in the sense of disproving Junk DNA.

Marc said...

It all depends on your definition of functional; one man's 'spurious binding sites' are another man's 'sites of protein sequestration'. Just because it does not have an obvious function at the moment does not mean it will never have a function as our knowledge of biological systems grows.

CuriousOnlooker said...

Dr. Moran, just curious: Why do you think your definition of "functional" is superior to Science's definition?

Diogenes said...

I'll answer that.

Your question is based on a false premise, and I think you calculated it that way.

Your question is based on the false premise that Moran thinks his definition of "function" is superior to Science's definition. He claimed no such thing. In biochemistry, "function" can have many, many, many definitions and each can be relevant in different contexts.

However, Science has claimed that ENCODE "functional" assays (actually assays for biochemical activity) disproved the Junk DNA hypothesis. If one is referencing the Junk DNA hypothesis, one must use the definition of "function" relevant to the Junk DNA hypothesis.

The definition of "function" invented wholecloth by Ewan Birney, function = biochemical activity, may be relevant in some contexts, but it is remotely distant from the definition relevant to the Junk DNA hypothesis.

Even way back in 1980, the scientists who coined the term "Junk DNA" explicitly said that much Junk DNA can be transcribed into RNA. So if one claims that RNA transcription could in principle disprove the Junk DNA hypothesis, they're either lying or very, very ignorant.

It is profoundly, profoundly dishonest to claim that ENCODE biochemical assays were capable of disproving the Junk DNA hypothesis, by engaging in equivocation-- switching between two definitions of a single word.

CuriousOnlooker, please be careful about asking questions based on false premises. If you do it again, I will start to get pissed off.

Anonymous said...

I am not Larry, yet ...

It's not Science's definition, but ENCODE's. This is not a matter of superiority but of meaning. That repetitive DNA could be transcribed accidentally once in a while, and thus be matched to many more regions than were actually transcribed, is not function proper, but transcriptional accident. That DNA can be remodelled, recombined, and such, because of some repetitive DNA regions is not function proper either, but shit that happens because of the nature of the recombination process itself. Et cetera.

But you don't care. Do you?

Faizal Ali said...

If I may: It's not really a question of which definition if "superior", but rather of whether the definition is being used consistently. If you use a definition of "functional" that would include most junk DNA, then go on to say since 80% of the genome is "functional" this means that 80% of the genome is not junk, you are obviously being inconsistent.

Larry Moran said...

Curious Onlooker asks,

Why do you think your definition of "functional" is superior to Science's definition?

I think that "function" has to mean a biologically relevant activity that affects the survival of the organism.

Let me give you an example. Some protein-encoding genes become inactivated by acquiring a mutation that causes premature termination of protein synthesis. This is a very well-known phenomenon in bacterial and 'phage genetics and it's also one of the mechanisms giving rise to pseudogenes in our genome.

Those pseudogenes can still have active promoters and they are still transcribed to produce mRNA. The promoter region still binds various transcription factors.

The gene is nonfunctional by my definition because it has lost its biological activity. It is now junk DNA.

However, this region of the genome is still "functional" by the definition used by ENCODE publicists and by Science editors because it is transcribed and because proteins still bind. I don't think such a re-definition of "functional" is particularly helpful in understanding how our genome works.

As others have pointed out, the re-definition of "functional" includes known examples of junk DNA so it really doesn't make any contribution to the debate over the amount of junk in our genome.

CuriousOnlooker said...

Diogenes: Haha! You really think I give a furry rodent's posterior about you getting pissed off? I don't because:

a. This is not your blog.
b. Whether you see it that way or not, I did feel that the definition of DNA having a protein-coding function was somehow seen as "superior". It's not my fault that you are incapable of comprehending a question as anything other than an ad hominem attack.

And by the way, it's perfectly fine to believe that one definition in science can be superior to others. For instance, the definition of complex life as evolved from simpler organisms is superior to the definition of life as being created by a supernatural being.

CuriousOnlooker said...

Thanks Dr. Moran; if I understand you correctly, what you mean is that "functional" should be defined as something that aids in the fitness of an organism's progeny.

And I think Diogenes could learn a thing or two from your reply.

Larry Moran said...

I have been a critic of many peer-reviewed papers for over 40 years.

I have never claimed that "peer-review" ensures accuracy. I subscribe to the idea that most of what's published in the scientific literature is wrong and will be completely forgotten a few months after publication.

It's creationists who have a problem with peer review. They categorically reject all those thousands of peer-reviewed papers supporting evolution but, when one paper raises doubts about a particular aspect of evolution, the creationists fall all over themselves praising it as genuine peer-reviewed science.

And whenever a creationist gets published in the peer-reviewed literature, they shout it from the rooftops. They want to have their cake and eat it too.

Diogenes said...

CuriousOnlooker:

Whether you see it that way or not, I did feel that the definition of DNA having a protein-coding function was somehow seen as "superior".

Jesus, we're back to the goddamn "non-coding DNA = junk DNA" bullshit. Scientists have fucking known since the 1960's that non-coding DNA can have a function, by the definition of function relevant to Junk DNA.

How many times do I have to list all the goddamn scientists who won goddamn Nobel Prizes from the goddamn Swedish committee for finding functions in non-coding DNA? How many times do I have to copy and paste that fucking list?

Nobel Prize for Jacques Monod and co-workers, 1965, for finding functions in non-coding DNA (regulatory elements).

Nobel Prize for Barbara McClintock in 1983 for her discovery of new functions in non-coding RNA (mobile genetic elements.)

Nobel Prize for Tom Cech and Sidney Altman in 1989, for discovery of catalytic functions resulting from non-coding DNA (catalytic RNA= ribozymes).

Nobel Prize for Jack Szostak and co-workers in 2009, for research in 1980’s on function in non-coding DNA (telomeres).

Nobel Prize for Richard Roberts and Phillip Sharp in 1993 for discovering introns (in non-coding DNA).

The structure of tRNA was known by 1964, crystal structure solved in 1974. tRNA is made from non-coding RNA.

The ribosome was known to be largely nucleic acid in the 1950's, general molecular structure known since the early 1970s, by the 1980's it was known the ribosome was a ribozyme-- based on functions residing in non-coding DNA.

Why do you think they got those Nobel Prizes? Because everybody has known since the 1960's that non-coding DNA can have a function, by the definition of function relevant to Junk DNA.

Blas said...

Seems that both you and creationists chooses which paper is science and wich not.

Mikkel Rumraket Rasmussen said...

There's nothing wrong with genuine criticism. As long as the "choosing" is based on evidential justification and scientific rigour, not preconcieved religious ideology.

When creationists "choose", they operate on the principle that the Bible is litterally true observations from the real world be damned. They've stated as much countless times.

Here's the evidence:
http://i.imgur.com/yrnaH.gif

You'll find no such statements constituting the consensus operating principles of science.

Creationism is mind disease.

gillt said...

Moran: "I think that "function" has to mean a biologically relevant activity that affects the survival of the organism."

I'm surprised you think fitness is important in a definition of functional genomics and I'm not sure how it's relevant. I think the way we understand function is in the relationship of genome to phenotype.

CuriousOnlooker said...

-Nobel Prize for Jacques Monod *and co-workers*.

Actually it was just two of them, Jacques Monod and Francois Jacob. Andre Lwoff got it for something different, for contributions to the goddamn theory of lysogeny.

Piotr Gąsiorowski said...

And by the way, it's perfectly fine to believe that one definition in science can be superior to others. For instance, the definition of complex life as evolved from simpler organisms is superior to the definition of life as being created by a supernatural being.

Neither of these is a definition. A definition explains the meaning of a term, not the origin of natural phenomena.

Diogenes said...

Actually it was just two of them, Jacques Monod and Francois Jacob. Andre Lwoff got it for something different, for contributions to the goddamn theory of lysogeny.

Hey, watch your mouth. There's no call for language like that.

Faizal Ali said...

Science publishes peer-reviewed research, but not everything it publishes is peer-reviewed research. A year end "Top 10" list falls into the "not peer-reviewed research" category.