More Recent Comments

Saturday, July 30, 2022

Wikipedia blocks any mention of junk DNA in the "Human genome" article

Wikipedia has an article on the Human genome. The introduction includes the following statement,

Human genomes include both protein-coding DNA genes and various types of DNA that does not encode proteins. The latter is a diverse category that includes DNA coding for non-translated RNA, such as that for ribosomal RNA, transfer RNA, ribozymes, small nuclear RNAs, and several types of regulatory RNAs. It also includes promoters and their associated gene-regulatory elements, DNA playing structural and replicatory roles, such as scaffolding regions, telomeres, centromeres, and origins of replication, plus large numbers of transposable elements, inserted viral DNA, non-functional pseudogenes and simple, highly-repetitive sequences.

This is a recent improvement (July 22, 2022) over the original statement that simply said, "Human genomes include both protein-coding DNA genes and noncoding DNA." I noted in the "talk" section" that there was no mention of junk DNA in the entire article on the human genome so I added a sentence to the end of the section quoted above. I said,

Some non-coding DNA is junk, such as pseudogenes, but there is no firm consensus over the total mount of junk DNA.1

This was immediately removed by a Wikipedia editor named Agricolae who seems to be more interested in history than science. His comment on removing my sentence was, "replace 'junk' DNA term, which is completely discredited as a useful characterization." I added it back and Agricolae removed it again saying, "'junk' is no longer considered scientifically valid or useful terminology." I suggested in the "talk" section of the article on the human genome that Agrocilae should read up on the convtroversy about junk DNA and supplied a link to my blog post on "Five Things You Should Know if You Want to Participate in the Junk DNA Debate. Then Agricolae added the following comments.

This is not the 1980s, when if DNA didn't encode proteins then we just threw up our hands and dismissively called it 'junk'. A promoter is not 'junk'. A centromere is not 'junk'. rDNA is not 'junk'. And most importantly, these don't belong in the same heterogeneous category with each other, let alone in the same artificial category with LINEs, pseudogenes, viral insertions, etc. We already mention so-called 'junk' DNA in the article, just not using the inaccurate term for it - "Human genomes include both protein-coding DNA genes and various types of DNA that does not encode proteins. The latter is a diverse category that includes DNA coding for non-translated RNA, such as that for ribosomal RNA, transfer RNA, ribozymes, small nuclear RNAs, and several types of regulatory RNAs. It also includes promoters and their associated gene-regulatory elements, DNA playing structural and replicatory roles, such as scaffolding regions, telomeres, centromeres, and origins of replication, plus large numbers of transposable elements, inserted viral DNA, non-functional pseudogenes and simple, highly-repetitive sequences." There is little benefit in ignoring this diversity, lumping it all together under a heterogeneous catch-all term of 'all of the various different things that aren't X'. More useful would be to break out the unhelpful '90%' figure into its distinct constituent parts, which data should be in the 2022 Science Special Issue (currently behind a paywall for me). Agricolae (talk) 20:57, 29 July 2022 (UTC)

Just a followup, elsewhere you have defined junk DNA as "DNA that can be deleted from the genome without affecting the fitness of the individual or the species". I would suggest that no piece of the human DNA has been characterized to the level at which we would know this for certain, and it would be unethical to carry out the test. It amounts to a 'we don't know of a benefit so there must not be one' argument. Given that pseudogenes can provide a venue for variation that can then be incorporated back into the transcriptome through gene conversion or reactiveation, that we all have at least one retrotransposed mRNAs in our genome that is a fully-functionaltranscribed protein because it randomly inserted next to a promoter, or a promoter randomly inserted next to it, and that LINEs allow for unequal crossing over and consequent gene duplication, specialization and diversification, I don't know how you point to a particular piece of DNA and say that having it can't possibly ever provide a fitness advantage. In your blog you point to the range of difference in genome size, but there are genomes stripped down to the point where they have next to no 'junk DNA' - if they can, so could we, but we don't - it has been argued that this is evidence for a functional benefit to all that misnamed 'junk', even if we don't know how such a fitness benefit might come about. And all of that doesn't get around the fundamental problem that 'junk' DNA is a heterogeneous category, defined by what it isn't, and like Tolstoy's unhappy families, all unhappy in their own different way, nothing more than a catchall for whatever things that are not something else, a combination of a linguistic holdover from a previous time and an expression of current imperfect knowledge and imagination. Even if it really is useless, and we certainly can't be sure that is the case, we are much better off talking individually about pseudogenes, LINEs, etc., rather than about an artificial category including all of these disparate things, each unhappy for its own particular reason. We are usually better off describing green, and blue, and yellow, rather than creating a page for 'colours that aren't red'.Agricolae (talk) 00:35, 30 July 2022 (UTC)

You will immediately recognize the flaw in the first comment. Agricolae doesn't know the difference between junk DNA and noncoding DNA and has fallen for the lie that all noncoding DNA was declared to be junk. The irony here is that Agricolae mentions all the functional nocoding regions that I have been posting about for the past thirty years. This includes scaffolding regions and origins of replication that hardly anyone else mentioned when I first began writing about this stuff. It's also clear that Agricolae didn't read my blog post on Five Things You Should Know if You Want to Participate in the Junk DNA Debate and has no idea that there's evidence for junk DNA. It's not just DNA that we don't know about.

This false notion of junk DNA permeates the second comment but it's compounded by some classic logical fallacies. Here's what I said to Agricolae on the talk thred.

Much of what you have just written is scientifically incorrect, misleading, or logically flawed (sometimes all three). But your lack of knowledge of the scientific controversy over junk DNA isn't really the point. The point is that there are many highly respected and intelligent scientists who say that most of our genome consists of junk DNA. You may not agree with them but you are wrong to use your power as a Wikipedia editor to impose your personal opinion on a Wikipedia article. It's ridiculous to censure any mention of "junk DNA" when the term is used in lots of other Wikipedia articles and is part of a very interesting and ongoing controversy about the content of the human genome.

"Wrong does not cease to be wrong because the majority share in it."

      Leo Tolstoy

This is a serious problem on Wikipedia. Many of the articles that interest me are protected by editors that jealously defend their personal views and prevent any dissent. Many of these editors are not critical thinkers and are not as knowledgeable about the subject as they think. They are very much interested in the general consensus view and they dismiss all minority views.


1. The article contains a short section on pseudogenes, describing them as "nonfunctional."

10 comments :

Athel Cornish-Bowden said...

As you will gather from our earlier discussion I agree with Larry about this. I have edited the Wikipedia article (restoring Larry's sentence), but probably it won't stay like that. I've also commented on the Talk page. If other contributors to this discussion are Wikipedia editors it would be a good idea for them to join in. (You don't actually need to be registered as an editor -- anyone can do it.)

Joe Felsenstein said...

I am with you and Athel Cornish-Bowden on this, and I'll try to make a comment on the Wikipedia Talk page too. I would say that 99% of people who actively work on molecular evolution conclude that more than 80% (nost, more than 90%) of the human genome is junk DNA. Unfortunately they are outnumbered by genomicists and molecular biologists who think that junk DNA is nearly nonexistent. It is the molecular evolutionists who understand the evidence, the others having been stampeded by poor arguments. It is very sad that Wikipedia will not be telling its readers about this very important divergence of views.
Joe Felsenstein, Professor Emeritus, Department of Genome Sciences, University of Washington, Seattle (also Professor Emeritus, Department of Biology). Former President, Society of Molecular Biology and Evolution, etc, etc.

Mikkel Rumraket Rasmussen said...

I left a comment too.

Athel Cornish-Bowden said...

As I feared, my edit didn't survive more than a few hours.

Matt G said...

These people are making the same mistake as the IDiots and "third way" evolutionists, and it's a teleological mistake: if it's there, it must have a function. Some have called junk DNA the dark matter of the genome, but this is unsupportable since we know its origins. They are ignoring prior probability, which puts the burden of proof even more squarely on them. If they are going to claim function for TEs, pseudogene, dead viruses, etc., whether SE or CE or any other definition, they have to demonstrate it.

Larry Moran said...

I added it back.

Athel Cornish-Bowden said...

Unfortunately Agricolae's User Page is completely blank (and always has been, apart from sentences occasionally added by vandals and then deleted), so we are completely in the dark about what expertise lies behind the opinions so dogmatical expressed. Did [s]he do a PhD with Francis Crick, or take a short course in biology at Bob Jones University? There is no way to know. [S]he has a huge number of edits (too many to count), nearly all of them concerned with such matters as British Israelism, slavery in Spain, obscure historical figures, etc., but virtually nothing relevant to the human genome. I think it's important to continue to demand serious references to support the dogmatism.

Matt G said...

Agricolae mentions deletion experiments being unethical in humans, but while perusing Larry's older material I came across a study in which large swaths were removed from the mouse genome with no obvious ill effects.

Larry Moran said...

The best definition of "function" is a sequence that can't be deleted without affecting the fitness of the organism. This is equivalent to saying that it is under purifying selection.

Agricolae doesn't object to talking about functional sequences in the Wikipedia article. It's strange that he doesn't want to define "function" and doesn't want to bring up the fact that you can't test for it by deleting human functional sequences.

But when you say that non-functional sequences (junk DNA) CAN be deleted without affecting fitness, he gets all uppity about definitions. Isn't that strange?

Josh Lang said...

There have been a few findings based on analysis of the non-coding regions of “junk DNA” in the human genome and they found non-random distributions that are similar to what we find with languages.

Based on these findings of non-randomness, and the very fact that we don't know what we don't know, I would absolutely not use the term "junk DNA". It's time to move on from the 1990's.