Nils Walter disputes junk DNA: (5) What does the number of transcripts per cell tell us about function?

I'm discussing a recent paper published by Nils Walter (Walter, 2024). He is arguing against junk DNA by claiming that the human genome contains large numbers of non-coding genes.

This is the fifth post in the series. The first one outlines the issues that led to the current paper and the second one describes Walter's view of a paradigm shift. The third post describes the differing views on how to define key terms such as 'gene' and 'function.' The fourth post makes the case that differing views on junk DNA are mainly due to philosophical disagreements.

-Nils Walter disputes junk DNA: (1) The surprise

-Nils Walter disputes junk DNA: (2) The paradigm shaft

-Nils Walter disputes junk DNA: (3) Defining 'gene' and 'function'

-Nils Walter disputes junk DNA: (4) Different views of non-functional transcripts

Transcripts vs junk DNA

The most important issue, according to Nils Walter, is whether the human genome contains huge numbers of genes for lncRNAs and other types of regulatory RNAs. He doesn't give us any indication of how many of these potential genes he thinks exist or what percentage of the genome they cover. This is important since he's arguing against junk DNA but we don't know how much junk he's willing to accept.

There are several hundred thousand transcripts in the RNA databases. Most of them are identified as lncRNAs because they are bigger than 200 bp. Let's assume, for the sake of argument, that 200,000 of these transcripts have a biologically relevant function and therefore there are 200,000 non-coding genes. A typical size might be 1000 bp so these genes would take up about 6.5% of the genome. That's about 10 times the number of protein-coding genes and more than 6 times the amount of coding DNA.

That's not going to make much of a difference in the junk DNA debate since proponents of junk DNA argue that 90% of the genome is junk and 10% is functional. All of those non-coding genes can be accommodated within the 10%.

The ENCODE researchers made a big deal out of pervasive transcription back in 2007 and again in 2012. We can quibble about the exact numbers but let's say that 80% of the human is transcribed. We know that protein-coding genes occupy at least 40% percent of the genome so much of this pervasive transcription is introns. If all of the presumptive regulatory genes are located in the remaining 40% (i.e. none in introns), and the average size is 1000 bp, then this could be about 1.24 million non-coding genes. Is this reasonable? Is this what Nils Walter is proposing?

I think there's some confusion about the difference between large numbers of functional transcripts and the bigger picture of how much total junk DNA there is in the human genome. I wish the opponents of junk DNA would commit to how much of the genome they think is functional and what evidence they have to support that position.

But they don't. So instead we're stuck with debates about how to decide whether some transcripts are functional or junk.

What does transcript concentration tell us about function?

If most detectable transcripts are due to spurious transcription of junk DNA then you would expect these transcripts to be present at very low levels. This turns out to be true as Nils Walter admits. He notes that "fewer than 1000 lncRNAs are present at greater than one copy per cell."

This is a problem for those who advocate that many of these low abundance transcripts must be functional. We are familiar with several of the ad hoc hypotheses that have been advanced to get around this problem. John Mattick has been promoting them for years [John Mattick's new paradigm shaft].

Walter advances two of these excuses. First, he says that a critical RNA may be present at an average of one molecule per cell but it might be abundant in just one specialized cell in the tissue. Furthermore, their expression might be transient so they can only be detected at certain times during development and we might not have assayed cells at the right time. I assume he's advocating that there might be a short burst of a large number of these extremely specialized regulatory RNAs in these special cells.

As far as I know, there aren't many examples of such specialized gene expression. You would need at least 100,000 examples in order to make a viable case for function.

His second argument is that many regulatory RNAs are restricted to the nucleus where they only need to bind to one regulatory sequence to carry out their function. This ignores the mass action laws that govern such interactions. If you apply the same reasoning to proteins then you would only need one lac repressor protein to shut down the lac operon in E. coli but we've known for 50 years that this doesn't work in spite of the fact that the lac repressor association constant shows that it is one of the tightest binding proteins known [DNA Binding Proteins]. This is covered in my biochemistry textbook on pages 650-651.¹

If you apply the same reasoning to mammalian regulatory proteins then it turns out that you need 10,000 transcription factor molecules per nucleus in order to ensure that a few specific sites are occupied. That's not only because of the chemistry of binary interactions but also because the human genome is full of spurious sites that resemble the target regulatory sequence [The Specificity of DNA Binding Proteins]. I cover this in my book in Chapter 8: "Noncoding Genes and Junk RNA" in the section titled "On the important properties of DNA-binding proteins" (pp. 200-204). I use the estrogen receptor as an example based on calculations that were done in the mid-1970s. The same principles apply to regulatory RNAs.

This is a disagreement based entirely on biochemistry and molecular biology. There aren't enough examples (evidence) to make the first argument convincing and the second argument makes no sense in light of what we know about the interactions between molecules inside of the cell (or nucleus).

Note: I can almost excuse the fact that Nils Walter ignores my book on junk DNA, my biochemistry textbook, and my blog posts, but I can't excuse the fact that his main arguments have been challenged repeatedly in the scientific literature. A good scientist should go out of their way to seek out objections to their views and address them directly.

---

1. In addition to the thermodynamic (equilibrium) problem, there's a kinetic problem. DNA binding proteins can find their binding sites relatively quickly by one dimensional diffusion—an option that's not readily available to regulatory RNAs [Slip Slidin' Along - How DNA Binding Proteins Find Their Target].

Walter, N.G. (2024) Are non‐protein coding RNAs junk or treasure? An attempt to explain and reconcile opposing viewpoints of whether the human genome is mostly transcribed into non‐functional or functional RNAs. BioEssays:2300201. [doi: 10.1002/bies.202300201]

Nils Walter disputes junk DNA: (4) Different views of non-functional transcripts

I'm discussing a recent paper published by Nils Walter (Walter, 2024). He is trying to explain the conflict between proponents of junk DNA and their opponents. His main focus is building a case for large numbers of non-coding genes.

This is the third post in the series. The first one outlines the issues that led to the current paper and the second one describes Walter's view of a paradigm shift. The third post describes the differing views on how to define key terms such as 'gene' and 'function.' In this post I'll describe the heart of the dispute according to Nils Walter.

-Nils Walter disputes junk DNA: (1) The surprise

-Nils Walter disputes junk DNA: (2) The paradigm shaft

-Nils Walter disputes junk DNA: (3) Defining 'gene' and 'function'

Nils Walter disputes junk DNA: (3) Defining 'gene' and 'function'

I'm discussing a recent paper published by Nils Walter (Walter, 2024). He is trying to explain the conflict between proponents of junk DNA and their opponents. His main focus is building a case for large numbers of non-coding genes.

This is the third post in the series. The first one outlines the issues that led to the current paper and the second one describes Walter's view of a paradigm shift.

-Nils Walter disputes junk DNA: (1) The surprise

-Nils Walter disputes junk DNA: (2) The paradigm shaft

Any serious debate requires some definitions and the debate over junk DNA is no exception. It's important that everyone is on the same page when using specific words and phrases. Nils Walter recognizes this so he begins his paper with a section called "Starting with the basics: Defining 'function' and 'gene'."

Nils Walter disputes junk DNA: (2) The paradigm shaft

I'm discussing a recent paper published by Nils Walter (Walter, 2024). He is trying to explain the conflict between proponents of junk DNA and their opponents. His main focus is building a case for large numbers of non-coding genes.

This is the second post in the series. The first one outlines the issues that led to the current paper.

Nils Walter disputes junk DNA: (1) The surprise

Walter begins his defense of function by outlining a "paradigm shift" that's illustrated in Figure 1.

FIGURE 1: Assessment of the information content of the human genome ∼20 years before (left)[110] and after (right)[111] the Human Genome Project was preliminarily completed, drawn roughly to scale.[9] This significant progress can be described per Thomas Kuhn as a “paradigm shift” flanked by extended periods of “normal science”, during which investigations are designed and results interpreted within the dominant conceptual frameworks of the sub-disciplines.[9] Others have characterized this leap in assigning newly discovered ncRNAs at least a rudimentary (elemental) biochemical activity and thus function as excessively optimistic, or Panglossian, since it partially extrapolates from the known to the unknown.[75] Adapted from Ref. [9].

Reference #9 is a paper by John Mattick promoting a "Kuhnian revolution" in molecular biology. I've already discussed that paper as an example of a paradigm shaft, which is defined as a strawman "paradigm" set up to make your work look like revolutionary [John Mattick's new paradigm shaft]. Here's the figure from the Mattick paper.

The Walter figure is another example of a paradigm shaft—not to be confused with a real paradigm shift.¹ Both pie charts misrepresent the amount of functional DNA since they don't show regulatory sequences, centromeres, telomeres, origins of replication, and SARS. Together, these account for more functional DNA than the functional regions of protein-coding genes and non-coding genes. We didn't know the exact amounts in 1980 but we sure knew they existed. I cover this in Chapter 5 of my book: "The Big Picture."

The 1980 view also implies, incorrectly, that we knew nothing about the non-functional component of the genome when, in fact, we knew by then that half of our genome was composed of transposon and viral sequences that were likely to be inactive, degenerate fragments of once active elements. (John Mattick's figure is better.)

The 2020 view implies that most intron sequences are functional since introns make up more than 40% of our genome but only about 3% of the pie chart. As far as I know, there's no evidence to support that claim. About 80% of the pie chart is devoted to transcripts identified as either small ncRNAs or lncRNAs. The implication is that the discovery of these RNAs represents a paradigm shift in our understanding of the genome.

The alternative explanation is that we've known since the late 1960s that most of the human genome is transcribed and that these transcripts—most of which turned out to be introns—are junk RNA that is confined to the nucleus and rapidly degraded. Advances in technology have enabled us to detect many examples of spurious transcripts that are present transiently at low levels in certain cells. I cover this in Chaper 8 of my book: "Noncoding Genes and Junk RNA.

The whole point of Nils Walter's paper is to defend the idea that most of these transcripts are functional and the alternative explanation is wrong. He's trying to present a balanced view of the controversy so he's well aware of the fact that some of us interpret the red part of the pie chart as spurious transcripts (junk RNA). If he's wrong, and I am right, then there's no paradigm shift.

You don't get to shift the paradigm all on our own, even if John Mattick is on your side. A true paradigm shift requires that the entire community of scientists changes their perspective and that hasn't happened.

In the next few posts we'll see whether Nils Walter can make a strong case that all those lncRNAs are functional. They cover about two-thirds of the genome in the pie chart. If we assume that the average length of these long transcripts is 2000 bp then this represents one million transcripts and potentially one million non-coding genes.

---

1. The term "paradigm shaft" was coined by reader Diogenes in a comment on this blog from many years ago.

Walter, N.G. (2024) Are non‐protein coding RNAs junk or treasure? An attempt to explain and reconcile opposing viewpoints of whether the human genome is mostly transcribed into non‐functional or functional RNAs. BioEssays:2300201. [doi: 10.1002/bies.202300201]

Nils Walter disputes junk DNA: (1) The surprise

Nils Walter attempts to present the case for a functional genome by reconciling opposing viewpoints. I address his criticisms of the junk DNA position and discuss his arguments in favor of large numbers of functional non-coding RNAs.

Nils Walter is Francis S. Collins Collegiate Professor of Chemistry, Biophysics, and Biological Chemistry at the University of Michigan in Ann Arbor (Michigan, USA). He works on human RNAs and claims that, "Over 75% of our genome encodes non-protein coding RNA molecules, compared with only <2% that encodes proteins." He recently published an article explaining why he opposes junk DNA.

Walter, N.G. (2024) Are non‐protein coding RNAs junk or treasure? An attempt to explain and reconcile opposing viewpoints of whether the human genome is mostly transcribed into non‐functional or functional RNAs. BioEssays:2300201. [doi: 10.1002/bies.202300201]

The human genome project's lasting legacies are the emerging insights into human physiology and disease, and the ascendance of biology as the dominant science of the 21st century. Sequencing revealed that >90% of the human genome is not coding for proteins, as originally thought, but rather is overwhelmingly transcribed into non-protein coding, or non-coding, RNAs (ncRNAs). This discovery initially led to the hypothesis that most genomic DNA is “junk”, a term still championed by some geneticists and evolutionary biologists. In contrast, molecular biologists and biochemists studying the vast number of transcripts produced from most of this genome “junk” often surmise that these ncRNAs have biological significance. What gives? This essay contrasts the two opposing, extant viewpoints, aiming to explain their basis, which arise from distinct reference frames of the underlying scientific disciplines. Finally, it aims to reconcile these divergent mindsets in hopes of stimulating synergy between scientific fields.

How to end the war in Ukraine according to a Canadian Conservative "diplomat"

In my opinion, the war in Ukraine is much more complicated than most people realize. We are constantly bombarded with propaganda from all sides and it inhibits rational thinking. One of the few reliable facts is that Vladimir Putin is a very smart bad person.

Lots of people think they have the answer to ending the war in Ukraine. One of the latest pundits is Chris Alexander who has published his thoughts in the Feb. 16, 2024 edition of Canada's Globe and Mail: Ukraine is paying the price for our nonchalance toward Russia’s leadership. Alexander spent years in Canada's Foreign Service, including many years in Moscow and a stint as Canada's ambassador to Afghanistan. In 2011 he was elected to Parliament as a Conservative MP and served as Minister of Citizenship and Immigration in Stephen Harper's cabinet. His reputation as a politician was very different than his previous, mostly admirable, reputation as a diplomat. Here's a excerpt from his Wikipedia article.

Copilot answers the question, "What is junk DNA?"

The Microsoft browser (Edge) has a built in function called Copilot. It's an AI assistant based on ChatGPT-4.

I decided to test it byt asking "What is junk DNA?" and here's the answer it gave me.

Older but wiser?

With age comes wisdom, but sometimes age comes alone.

Oscar Wilde

Like many baby boomers, I sometimes forget people's names and other important bits of information. Sometimes I can't find a word that's been in my vocabulary for decades. These lapses are often temporary but very annoying. It's a sign of age. (I am 77 years old.)

We often make fun of these incidents and consol ourselves with the knowledge that we may be old but we are much wiser than we were in our younger days. We have years and years of experience behind us and over the years we've learned a thing or two that we never understood when we were listening to the Beatles on the radio. We've lived through the Cuban Missile crisis, the war in Viet Nam, the assassination of two Kennedys and Martin Luther King, and a host of cultural changes. We've lived in several different countries and we've raised children. All of these experiences have made us wiser, or so we think.

Open and closed chromatin domains (and epigenetics)

Gene expression in eukaryotes is influenced by the state of chromatin. Tightly packed nucleosomes inhibit the binding of transcription factors and RNA polymerase so that genes in these regions are "repressed." From time to time these regions loosen up a bit allowing access to transcription complexes and subsequent transcription.

The tightly packed regions are known as closed domains and the accessible regions are open domains. Some authors add an intermediate domain called a permissive domain. This model of eukaryotic gene expression has been around for 50 years and the important mechanisms controlling the switch were worked out in the 1980s. I found a recent review that covers this issue in the context of epigenetics and the image below comes from that paper (Klemm et al., 2019).

Philip Ball's new book: "How Life Works"

Philip Ball has just published a new book "How Life Works." The subtitle is "A User’s Guide to the New Biology" and that should tell you all you need to know. This is going to be a book about how human genomics has changed everything.

"People also ask" about junk DNA

I'm interested in the spread of science misinformation on the internet. The misinformation about the human genome is a good example that illustrates the problem. There are many other examples but I happen to know a lot about this particular one.

Anyone trying to find out about junk DNA will find it impossible to get a correct answer by searching the internet. The correct answer is that the amount of junk DNA in the human genome is controversial: some scientists think that most of our genome is functional while others think that as much as 90% is junk. The scientific evidence strongly favors the junk side of the controvesy and that's very well explained in the Wikipedia articles on Junk DNA and Non-coding DNA.

Creation Myth: Biologists Thought All Non-Coding DNA was Junk

Watch this short video from Daniel Stern Cardinale. He read my book!

---

Benjamin Lewin's new book and his view of the human genome

I was a big fan of Benjamin Lewin. Back in the 1970's he published the first volumes of what was to become Genes, the authoritative textbook of molecular biology. I admired his ability to understand the latest experiments and put the results in the appropriate context.

Later on, when he founded the journal Cell, his editorials and other writings were always insightful. His editorial judgement was impeccable—he always published the very best papers in molecular biology.¹

Why do Intelligent Design Creationists lie about junk DNA?

A recent post on Evolution News (sic) promotes a a new podcast: Casey Luskin on Junk DNA’s “Kuhnian Paradigm Shift”. You can listen to the podcast here but most Sandwalk readers won't bother because they've heard it all before. [see Paradigm shifting.]

Luskin repeats the now familiar refrain of claiming that scientists used to think that all non-coding DNA was junk. Then he goes on to list recent discoveries showing that some of this non-coding DNA is functional. The truth is that no knowledgeable scientist ever claimed that all non-coding DNA was junk. The original idea of junk DNA was based on evidence that only 10% of the genome is functional and these scientists knew that coding regions occupied only a few percent. Thus, right from the beginning, the experts on genome evolution knew about all sorts of functional non-coding DNA such as regulatory sequences, non-coding genes, and other things.

Kat Arney interviews me on her podcast

I had a long chat with Kat Arney a few weeks ago and she has now taken the best parts of that conversation and put them in her latest Genetics Society podcast: Genes, junk and the 'dark genome'. My comments are in the last twelve minutes. At the end, Kat asks me "Is there like one thing you would really want a student or researcher, working in genetics today to really understand about the human genome?"

Kat was kind enough to write a blurb for my book last year where she said,

What's in Your Genome? is a thought-provoking and pugnatious book that will make you wonder afresh at the molecular intracies of life. When it comes to our genomes, we humans are nothing special—Moran makes a convincing argument that the vast majority of our sloppy human genome is not mysterious genetic treasures but boring junk.

In this podscast, she combines my thoughts on the human genome with those of two people who don't agee with the idea that the human genome is full of junk. Here's a brief summary of their positions.

Naomi Allen is Chief Scientist at UK Biobank, a consortium that's sequencing the genomes of UK citizens. So far, they've published data on 500,000 genome sequences. I wrote about one of their more significant findings last year (August, 2022) where they reported on the fraction of the human genome that was under purifying selection. This is an excellent proxy for functional DNA and the results are in line with (my) expectations: less that 10% of the genome is conserved and most of it is in the non-coding fraction [Identifying functional DNA (and junk) by purifying selection.

It's too bad that Kat's interview with Naomi Allen doesn't mention that important result, especially since the podcast is about junk DNA. Here's how Naomi Allen begins her part of the interview.

Whole genome sequencing enables researchers to look at all of the genetic variation across the entire genome. So not just in the 2% of the genome that encodes for proteins, but all of the genetic variation, much of which was previously considered "junk DNA" precisely because we didn't know what it did.

This is disappointing for two important reasons. First, surely in 2023 we've gone beyond the tired myth that all of the information in the human genome was concentrated in coding DNA? Second, no knowledgeable scientist ever said that all non-coding DNA was junk DNA and the idea of junk DNA was not based on ignorance so surely it's time to stop repeating that myth as well.

The rest of that interview focuses on how mapping genetic variation could contribute to our understanding of health and disease. I would have loved to ask how Biobanks proposes to do this if most of the variation is in junk DNA and also ask whether mutations in junk DNA can contribute to genetic disease. (They can.)

Danuta Jeziorska is the CEO of Nucleome Therapeutics, a company that's described as "spun out of Oxford University with a new set of technologies for exploring the dark genome." Kat asks her about the dark genome and here's her response.

So if you think about it, we have 22,000 genes in our genome, and we can compare that to having 22,000 ingredients in the fridge. We use the same set of ingredients to create different meals, just like how we have the same DNA within each cell, but then we have hundreds of different cell types. So this dark genome determines the combination of ingredients of the genes that you take and at which level you use them, to produce the different cell types that build our body. And you can just imagine that if you make a mistake in that - so let's say that you add the wrong ingredients in the wrong meal, you can mess up the meal. And in this same way you can mess up the cell type. So if you, for example, if you don't produce enough of haemoglobin to transport oxygen around the body, you will end up with a genetic form of anaemia or if you turn on a gene that's not supposed to be turned on, like an oncogene, you may end up having cancer.

So the dark genome is now very well understood as the mechanism that is causing diseases.

This is a slightly different definition of the dark genome than those I discussed in a recent post [What is the "dark matter of the genome"?]. In that post I suggested that most scientists were referring to all of the functions in non-coding DNA but Danuta Jeziorska seems to be restricting her use of "dark genome" to just regulatory sequences. In the rest of the interview she goes on to describe various types of regulatory sequences, with an emphasis on 3D structure, and to explain that many common genetic diseases are caused by mutations in regulatory sequences. Her company is using machine learning to find the functional elements in the dark genome and which variants are associated with disease. They are also investing in drug discovery.

---