In order to build the case for revolution, he tries to demonstrate a paradigm shift in our view of molecular biology by showing a huge gap between the understanding of previous generations of molecular biologists and the post-genomic view. I believe he is wrong about this for two reasons: first, he misrepresents the views of older molecular biologists and, second he misrepresents the discoveries of the past twenty years. I tried to explain why he was wrong about these two claims in a previous post where I discussed an article he published in Scientific American in May 2024: Philip Ball says RNA may rule our genome.
Philip Ball responded to my criticism in a comment under that article.
Older molecular biologists were really stupid
I said ...
Ball begins with the same old myth that writers like him have been repeating for many years. He claims that before ENCODE most molecular biologists were really stupid. According to Philip Ball, most of us thought that coding DNA was the only functional part of the genome and most of the rest was junk DNA.
In the comment section of my earlier post, Philip Ball says,
I’m sorry to say that Larry’s commentary here is dismayingly inaccurate.
Let’s get this one out of the way first:
“He claims that before ENCODE most molecular biologists were really stupid.”
I have never made this claim and never would – it is a pure fabrication on Larry’s part. I guess this is what John Horgan meant in his comment to Larry: credible writers don’t just make up stuff.
I admit that Philip Ball never said those exact words. I'll leave it to the readers to decide whether my characterization of his position is accurate.
I stand by the statements I made although I admit to a bit of hyperbole. Ball has said repeatedly that the molecular biologists of my generation were wedded to the idea that coding regions were the only important part of the genome and he often connects that to the Central Dogma of Molecular Biology. He also claims that the experts in molecular biology dismissed all non-coding DNA as junk. Here's how he puts it in another article that he published recently in Aeon: We are not machines.
Only around 1-2 per cent of the entire human genome actually consists of protein-coding genes. The remainder was long thought to be mostly junk: meaningless sequences accumulated over the course of evolution. But at least some of that non-coding genome is now known to be involved in regulating genes: altering, activating or suppressing their transcription in RNA and translation into proteins.
I interpret that to mean that older molecular biologists, like me, didn't know about functional non-coding DNAs such as centromeres, telomeres, origins of replication, non-coding genes, SARs, and regulatory sequences in spite of the fact that thousands of papers on these sequences were published in the 30 years that preceded the publication of the first draft of the human genome sequence. This is not true, we did know about those things. I don't think it's too much of an exaggeration to say that Philip Ball thinks we were really stupid.
Here's what he says in his book, "How Life Works" (p. 85) when he's talking about the beginning of the human genome project.
Even at its outset, it faced the somewhat troubling issue that just 2 percent or so of our genome actually accounts for protein-coding genes. The conventional narrative was that our biology was all about proteins, for each of which the genome held the template. ... But we had all this other DNA too! What was it for? The common view was that it was mostly just junk, like the stuff in our attics: meaningless material accumulated during evolution, which our cells had no motivation to clear out.
Again, his claim is that in 1990 at the beginning of the human genome project the experts in molecular biology thought that non-coding DNA was mostly junk (98% of the genome). I have repeatedly refuted this myth and challenged anyone to come up with a single scientific paper arguing that all non-coding DNA is junk. I challenge Philip Ball to find a single molecular biology textbook written before 1990 that fails to discuss regulation, non-coding genes, and other non-coding functional elements in the human genome.
The truth is that the molecular biology experts concluded in the 1970s that we had about 30,000 genes and that 90% of our genome is junk and 10% is functional. That 10% consisted of about 2% coding DNA (now thought to be only 1%) and 8% functional non-coding DNA. So the "conventional narrative" was that there was a lot more functional non-coding DNA than coding DNA.
The human genome is full of genes for regulatory RNAs.
"Ball is one of the most meticulous, precise science writers out there. He is the antithesis of hypey, "dumb-it-down" reporting. He is MUCH more credible than you are, Laurence."
John Horgan July, 2024The title of the article I was discussing is "Revolutionary Genetics Research Shows RNA May Rule Our Genome." In that article Ball says that ENCODE was basically right and there are many more non-coding genes than protein-coding genes. I pointed out that Ball mentions some criticism of this idea but only to dismiss it. I said that "[Ball] wants you to believe that almost of all of those transcripts are functional—that's the revolution that he's promoting." Philip Ball objects to this statement ...
This too is sheer fabrication. I don’t say this in my article, nor in my book. Instead, I say pretty much what Larry seems to want me to say, but for some reason he will not admit it – which is that there is controversy about how many of the transcripts are functional."
Ball states that "ENCODE was basically right" when they claimed that 75% of our genome was transcribed and he goes on to say that ...
Dozens of other research groups, scoping out activity along the human genome, also have found that much of our DNA is churning out 'noncoding' RNA.
He says that ENCODE has identified 37,000 noncoding genes but there may be as many as 96,000. After making these definitive statements, he mentions that there are "still doubters" but then discuss why these discoveries are revolutionary. Later on he quotes John Mattick suspecting that there may be more that 500,000 non-coding genes.
Toward the end of the article, after discussing all kinds of functional RNAs, he brings up the Ponting and Haerty review where they say that most lncRNAs are just noise. He also mentions that the low copy number of non-coding RNAs raises questions about whether they are functional but immediately counters with the standard excuses from his allies.
Ball closes the article with ...
Gingeras says he is perplexed by ongoing claims that ncRNAs are merely noise or junk, as evidence is mounting that they do many things. "It is puzzling why there is such an effort to persuade colleagues to move from a sense of interest and curiosity in the ncRNA field to a more dubious and critical one," he says.
Perhaps the arguments are so intense because they undercut the way we think our biology works. Ever since the epochal discovery about DNA's double helix and how it encodes information, the bedrock idea of molecular biology has been that there are precisely encoded instructions that program specific molecules for particular tasks. But ncRNAs seem to point to a fuzzier, more collective, logic to life. It is a logic that is harder to discern and harder to understand. ut if scientists can learn to live with the fuzziness, this view of life may turn out to be more complete.
What's remarkable about the quote from a leading ENCODE worker (Gingeras) is that he is "puzzled" by scientists who are dubious and critical about claims in the ncRNA field. Isn't that what good scientists are supposed to do? Isn't that exactly what we did when we successfully challenged the dubious claims about junk DNA made in 2012?
There is no doubt in my mind that Philip Ball has fallen hook-line-and-sinker for the ENCODE claims that our genome is buzzing with non-coding genes. He only brings up the counter-arguments to dismiss them and pretend that he is being fair. Nobody who was truly skeptical about the function of transcripts would write an article with the title, "Revolutionary Genetics Research Shows RNA May Rule Our Genome."
However, as Ball points out in other comments, he does have a sentence in his book where he mentions that perhaps only 30% of the genome is functional. He says in the comment that what he believes is that the amount of functional DNA lies somewhere between 10% and 30%. That's not something that he mentions in the Scientific American article but, if he's being honest, it does mean that I was unfair when I said he believes that "almost of all of those transcripts are functional" but I only know that from what he now says, not from the published article.
If I were to take Philip Ball at his word—as expressed in the comment—then he must believe that most of the ENCODE transcripts are junk RNA. That's not a belief that you get from reading his published work.2 Furthermore, if I were to take him at his word, then he must believe that there are some reasonable criteria that must be applied to a transcript in order to decide whether it has a biologically relevant function. So, when he says that ENCODE identified 37,600 non-coding genes he must have these criteria in mind but he doesn't express any serious skepticism about that number. We all know that there's no solid evidence that such a large number of transcripts are functional but that doesn't bother Philip Ball. He thinks we are in the middle of an RNA revolution.
1. In commenting to my previous post, Ball says he believes that somewhere between 70% and 90% of our genome is junk but he doesn't say this in the Scientific American article. Instead, he says that scientists were surprised to learn that 75% of the human genome is transcribed implying that there's a lot of function. He goes on the say that "ENCODE was basically right." But what the ENCODE publicity campaign actually said was that junk DNA is dead and there's practically no junk DNA. If Ball really believes that up to 90% of the genome is junk then to me this means that ENCODE was spectacularly wrong not "basically right."
2. Ball says that 75% of the genome is transcribed. If Ball believes that as little as 10% may be functional then he must believe that less than 10% is transcribed to produce functional RNAs since he has to allow for regulatory sequences and other functional DNA elements. Let's say that 8% is a reasonable number. Ball seems to be willing to admit that 67% of the genome might be transcribed to produce junk RNA.
8 comments :
In the early 1970s the Cambridge laboratory where I had worked described the power of dsRNA to inhibit protein synthesis as a probable mechanism of defence against intracellular pathogens. This is now widely recognized. By the same token, organisms must adapt their transcriptional repertoires to avoid self RNA species interacting to form dsRNA (which would tend to negatively select them).
So, the transgenerational utility of a section of DNA that would hybridize with the nucleic acid of intracellular pathogens was readily apparent. This might not be of selective value in an individual generation, but over many generations, from time-to-time, it would be selectively advantageous, and so retain its place in the genome.
Thus, in terms of the regular phenotype, it would appear neutral, but would be favoured transgenerationally. By the same token, we have multiple V region immunoglobulin genes. If there is a defect in one of these, one's line might survive for some generations quite well. I have contributed this case to the Sandwalk blog for some decades, sometimes in poetical form. For more details see my textbook (Evolutionary Bioinformatics, 3rd ed. 2016).
I’m going to let Larry’s “OK, he didn’t say what I said he said, but I reckoned he thought it so I wrote that he said it” stand testament to his intellectual and scholarly rigour. That applies equally to “He didn’t say what he believed about the magnitude of quantity X, so what else could I do but make up the number I thought he believed?”
Anyway, the mischaracterization continues. “[Ball] also claims that the experts in molecular biology dismissed all non-coding DNA as junk.” To support that claim, Larry quotes this from me: “Only around 1-2 per cent of the entire human genome actually consists of protein-coding genes. The remainder was long thought to be mostly junk.” I’m not going to insult the reader’s intelligence by pointing out how these two statements are not at all the same. (And after all, Larry seems to have dedicated his time to defending the idea that the remainder mostly *is* junk. And indeed, if by “most” we mean “the majority”, I totally accept that as a possibility.)
(Later this elision is repeated: “Again, his claim is that in 1990 at the beginning of the human genome project the experts in molecular biology thought that non-coding DNA was mostly junk (98% of the genome). I have repeatedly refuted this myth and challenged anyone to come up with a single scientific paper arguing that all non-coding DNA is junk.” Do I *really* have to says that “Mostly” does not mean “all”? I wrote “mostly” because I meant “mostly”, and indeed that seems to be the very position Larry takes: “molecular biology experts concluded in the 1970s that… 90% of our genome is junk”.)
Then: “I don't think it's too much of an exaggeration to say that Philip Ball thinks we were really stupid.” I guess this brings us to a dead end: I have stated that I don’t think that at all, so Larry believes I am simply lying. There is not much one can do when faced with that accusation, except to point out how rude and unprofessional it is.
[More follows below]
[Continued] Larry also wonders if I’m being honest about my guess about how much ncRNA is functional (why would I not be honest?) –
“if he's being honest, it does mean that I was unfair when I said he believes that "almost of all of those transcripts are functional" but I only know that from what he now says, not from the published article.”
Well, he might have known it too if he had read my book, and because he has so repeatedly disparaged it, I suppose I assumed he had done so. But I guess not, it seems. In any case, as I point out above, the Sci Am article didn’t indicate at all what I thought about the proportion of the genome that is functional – so why make blind guesses? The article didn’t indicate it because it isn’t especially relevant to the matter at hand, which is that we keep finding noncoding RNA doing stuff in the cell.
Perhaps the bigger issue that needs addressing is the unspoken narrative here of “Who’re you going to believe, an ex-physicist science writer or a professor of biochemistry?” I write what I write not because I figure I know more than know-nothing molecular biologists, but precisely because I speak to those of them who are actually studying these problems. So, for example, we have Larry saying that discoveries of non-coding RNA in the late 80s and 90s surprised no one who knew what they were talking about, versus, say, Nobel laureate Victor Ambros, who told me that it seemed so peculiar to find that lin4 encoded this tiny ncRNA in C. elegans that they thought it must be just a quirk of the worm. Or Jeanne Lawrence, who worked on one of the first known functional long ncRNAs (XIST) in the 90s:
Me: “Was it a shock that RNA could do this?
Jeanne: “It was totally surprising.”
Or we have geneticist Adrian Bird, who I consulted for my Sci Am piece precisely because I knew he’d expressed some scepticism about the functionality of all these ncRNAs (and also because I know he’s a very smart guy). Adrian said: ““Knowing about all these diverse roles for ncRNA transcription has clearly enhanced our understanding of gene control, and might even be classified as a paradigm shift”. (I’m not a fan of the P-word, but there you have it.)
But more importantly even than this, Ambros and veteran bioinformatics expert Peter Stadler both made the point that the surprise wasn’t just at the discovery of RNAs doing this stuff – it was surprise that the nice narrative already in place wasn’t the whole story. Here’s Victor:
“There was no theoretical need for these things. We knew transcription factors worked. We also knew there are lots of ways of regulating the activity of a protein. There’s no theoretical need for a small RNA.”
And here’s Peter:
“Mostly "everybody" believed that elucidating the function of all proteins will effectively explain how the cell works. Essentially, the model of transcription regulated by transcription factors [+ enhancers] was conceptually sufficient to explain how it is supposed to work. The discovery of non-coding RNAs was decided NOT a consequence of there being a missing piece in the explanatory paradigm. The community "knew" how the cell works. The discovery of microRNAs and the idea of post-transcriptional regulation, I think, was more of an annoyance, because it made life for complicated, rather than a welcome solution to open questions.”
So, readers here can choose to believe Larry when he says, without references, that none of this was any surprise to *him* (and indeed I’m quite prepared to believe that’s true, since he’s said it), but I hope they will be a little sceptical about the suggestion that this is what all molecular biologists thought and that it’s only the idiot science writer who’s trying to spin a different narrative. *That* aspect of Larry’s comments, at least, comes close to misinformation.
[Continued]
I don’t dismiss, in my article, critics of the idea that there is a lot of functional ncRNA. I report those criticisms, and give the advocates a chance to respond to them, leaving the reader to judge. That is normal journalism, which Larry seems not to understand – it rather feels as though he sees it either as “good” (if he agrees with it) or “bad” (if he doesn’t).
As for this: “I challenge Philip Ball to find a single molecular biology textbook written before 1990 that fails to discuss regulation, non-coding genes, and other non-coding functional elements in the human genome.” - for anyone tempted to believe Larry’s fantasy version of what textbooks in the 1980s and 90s said, I encourage them to look at the Ur-text, Alberts’ Molecular Biology of the Cell, 4th edition in *2002*. There, very little is said about noncoding RNA, and most of what is concerns the well-known ribosomal RNA and transfer RNA (which I say in my article were long known to biologists). Other than that, the only mention is to snoRNA and snRNA in the nucleolus and Cajal bodies. There’s no mention of microRNAs. There is no mention of long ncRNA as a distinct category, although XIST RNA is mentioned as an “unusual RNA molecule”. There is no mention of RNA whatsoever in the chapter on gene regulation, which is presented entirely in terms of protein transcription factors (and cofactors). So, essentially it’s the picture I give in the Scientific American article for what was known or thought around those times. (Of course there are also plenty of non-transcribed regulatory elements in the genome: promoters and so forth. If Larry wants to imply I’m somehow suppressing molecular biologists’ knowledge of these – well, I leave that for the reader to judge.)
I wrote my book, as well as pieces like this, not to suggest that molecular biologists have been ignorant fools, but for precisely the opposite reason: to show what exciting new understanding they are revealing. Which is of course science as usual. Larry seems to feel that suggesting there has been any significant novelty in understanding “how life works” in the past several decades is a slight to him and his colleagues. I find that peculiar. No one said, when dark energy (or whatever corresponds to it) was discovered in the 1990s, that physicists had been blinkered fools until then. They said “Wow, science!” I have stacks of quotes from informed biologists along the lines of “We thought we knew how this process worked, but now we’re having to think again.” They seem excited (if also daunted) by that, and so am I. So I’m not going to worry about responding to every insult or misquote that Larry might subsequently put up here, but am going to get on with the job of reporting the science.
There are three separate issues here.
1. Does Philip Ball try to convey to his readers the message that old molecular biologists didn't know about functional non-coding DNA such as regulatory sequences and non-coding genes? I'm not talking about quibbles over whether the word "mostly" appears in his writings. I'm talking about the general theme of his recent articles.
2. Does Philip Ball try to convince his readers that the human genome contains many more non-coding genes than protein-coding genes and does he do an adequate job of telling them about scientists who disagree with this claim? Does he ever seriously discuss what criteria we should use to establish the truth of those claims of functional genes?
3. Does Philip Ball tend to quote scientists who agree with his views much more favorably than scientists who disagree? Does he tend to leave his readers with the view that he is merely presenting the consensus view of post-genomic molecular biology? Does he do a good job of telling his readers that the ENCODE researchers were dead wrong in 2012 when they claimed that 80% of the genome is functional? Does he explain to his readers why they were wrong and why dozens of experts attacked those claims within days of the publication?
Here's an example of Philip Ball's writing from page 75 of his book "How Life Works: A User's Guide to the New Biology." You can judge for yourselves whether this is an accurate view of the state of molecular biology in the 1990s.
"THE PROBLEM WITH GENES"
"Despite all these caveats, complications and conditions, Crick's point remained: a gene was a means of getting a protein. But what then? The prevailing view up to the 1980s and 90s was that somehow the activation - the technical term is expression - of a gene to make a protein leads to some specific physiological or developmental outcome, and that there is some fairly direct link between genotype and phenotype, and a mapping of one to the other. Certain alleles should lead predictably to certain traits."
"As an editor at the science journal Nature in the 1990s, I began to sense that it wasn't so straightforward. I felt a bit like Keanu Reeve's character Neo in The Matrix when he starts to doubt that he can take the world around him at face value because of little inconsistencies and oddities: glitches that he is tempted to shrug off, but which once confronted, will compel a complete revision of how he thinks things work. In our weekly editorial meetings, my colleagues in the biological sciences team would regularly announce that they had accepted for publication a paper showing that gene X, thought to be involved in function Y - it might be limb development or pregnancy - turned out to be important too for phenomenon Z, such as cancer or the immune response. I was an editor for the physical sciences and a biological naïf, and would listen with great interest but minimal understanding. Finally I built up the courage to ask, 'So what does that mean?' The answer was delivered almost gleefully, 'We have no idea.'"
"It's no wonder my fellow editors were perplexed. Such observations didn't seem to fit with the consensus intuition that genes have specific roles to play in how cells and organisms form and function."
I maintain that after the groundbreaking work on Drosophila in the 1980s we had a pretty good idea about the role of developmentally regulated genes - mostly transcription factors - and how mutations in those genes could produce strange and pleiotropic effects. We may not have been able to immediately interpret each and every new example that was published but that didn't mean that we were on the verge of discovering that we lived in a matrix.
I get the impression from the quote from Philip Ball about the results that surprised him and his fellow editors, that pleiotropy was a new idea to them. It was certainly an old concept when I was taught about it in the 1960s. Checking Wikipedia I find that the word was available since 1910 and it was a matter of controversy until Beadle and Tatum's 1941 paper. I am surprised that the editors were surprised.
They were surprised that a transcription factor expressed at different times and in different locations can have different roles?
Post a Comment