- Genetic Load
Every newborn human baby has about 100 mutations not found in either parent. If most of our genome contained functional sequence information, then this would be an intolerable genetic load. Only a small percentage of our genome can contain important sequence information suggesting strongly that most of our genome is junk. - C-Value Paradox
A comparison of genomes from closely related species shows that genome size can vary by a factor of ten or more. The only reasonable explanation is that most of the DNA in the larger genomes is junk. - Modern Evolutionary Theory
Nothing in biology makes sense except in the light of population genetics. The modern understanding of evolution is perfectly consistent with the presence of large amounts of junk DNA in a genome. - Pseudogenes and broken genes are junk
More than half of our genomes consists of pseudogenes, including broken transposons and bits and pieces of transposons. A few may have secondarily acquired a function but, to a first approximation, broken genes are junk. - Most of the genome is not conserved
Most of the DNA sequences in large genomes is not conserved. These sequences diverge at a rate consistent with fixation of neutral alleles by random genetic drift. This strongly suggests that it does not have a function although one can't rule out some unknown function that doesn't depend on sequence.
Thursday, July 04, 2013
Five Things You Should Know if You Want to Participate in the Junk DNA Debate
Here are five things you should know if you want to engage in a legitimate scientific discussion about the amount of junk DNA in a genome.
6. Naive adaptationism is wrong, you can't just blithely assume everything is an adaptation.
ReplyDelete7. Contrary to your intuition, the energetic cost of replicating a large genome is very small compared to other energy expenditures in large eukaryote cells (e.g. muscle action).
Both of those are included in #3.
DeleteTrue, but lots of people reason from "having that much DNA must be really energetically costly" to "therefore it wouldn't persist unless it was functional". This would be a reasonable (though not unassailable) argument, but it depends on a false fact.
DeleteWhat about 7 in a protozoa? Is the cost of replicating the DNA significativly differnt to give an reproduction advantage?
DeleteWhy can't the 'junk' be assumed to be 3-D spacers so modifers for a gene have a close spacial relationship?
ReplyDeleteonion test
DeleteNevertheless #5 is problematic. Sequence non-conservation does not mean junk. Some segments could have non-conserved sequences yet serve as spacers.
DeleteLou Jost
Please explain which part of #5 is troubling you. Is it the part where I mention function that doesn't depend on sequence?
DeleteThe evidence that most of our genome is useless "junk" DNA seems convincing to me. I enjoy debating the topic with friends (taking the "Yes, it's junk!" side), and in fact did so a couple days ago, in a bog while recording plant diversity.
ReplyDeleteHowever, I find the "C-value paradox" less convincing that other lines of evidence. The C-value paradox is used as a kind of "Wow! Why would anything need 10 copies of every gene?!" argument that may not be more logical than the "No! Humans can't be related to chimps!" argument against evolution. That's sad, because polyploidy could be used better.
I think of a diploid hypothetical with 20 made-up units of DNA and a tetraploid with 40 units of DNA. Both survive well and are similar. The tetraploid as 4 copies of the triosphospate isomerase gene, twice as many as the diploid, but all four are functional. They aren't junk, how ever much we humans may think the tetraploid doesn't need more than two copies. Perhaps the tetraploid manages to get all its needed transcription done a little faster than the diploid, which may be why a higher proportion of plants are polyploid in high latitude and high elevation habitats where cold and short growing season combine to make completing growth and reproduction in one season difficult. Even if there is no speed advantage, those four copies are in use. In our hypothetical species pair, the proportion of functional and junk DNA (whatever it is) is the same in diploids and tetraploids. Polyploidy, C-value, itself didn't help make the junk DNA case.
It seems to me that polyploidy can provide some firmer lines of evidence for junk DNA.
First, high polyploids don't have even multiples of the amount of DNA in diploids. If in our hypothetical taxon, diploids have 20 units of DNA and tetraploids have 40, then hexaploids may have 58 and decaploids have 73. Conserving "balanced" amounts of DNA is unimportant. If some DNA can be jettisoned in high polyploids, was it unnecessary to begin with? (Argument #5 above.)
Second, the multiple copies of a functional gene aren't necessarily equally functional. In my isozyme work on polyploids (mostly tetraploids), I observed that a common pattern in tetraploids was as follows. For a given enzyme, one gene seemed to produce an invariant product, allele A. The homeologous gene pair seemed to produce alleles A and B, or often just alleles B, C, and N (null; non-functioning under the experimental conditions). This is part of how polyploidy produces wonderful variation for natural selection and genetic drift to act upon, but it is also evidence that in polyploids formerly functional genes can become non-functional junk. (Argument #5 again.)
So it seems to me that the C-value paradox boils down to a "Wow!" argument that may be helpful in winning debates but isn't as fully logical as we'd like, or a variation of argument #5 about unconserved sequences.
The c-value refers to the DNA content of haploid genomes.
DeleteI.e., the c-value of human haploid sperm and egg cells is the same as the c-value of diploid fibroblsts as well as tetra- and octoploid liver cells.
Hmmm. How does this play out in different species, some of them polyploid? It seems to me that it would work this way.
DeleteDiploid species, C-value = 10 units. Sperm and egg each have 10 units of DNA. Most body cells have 20 units. Tetra- and octoploid liver cells have 40 and 80 units, respectively. All these cells have the same C-value.
Closely related tetraploid species. C-value = 20 units. Sperm and egg have 20 each have 20 units of DNA. Most body cells have 40 units. Tetra- and octoploid liver cells have 80 and 160 units, respectively. All these cells have the same C-value, which is twice as much as that of the related diploid species.
If I'm wrong, I'm absolutely certain that will be explained to me here.
It also seems to me that when a tetraploid species evolves from a diploid ancestor (and speciation may be essentially instantaneous in these cases), the proportion of functional DNA (likely small) and junk DNA (likely high) will initially be the same in both the diploid and the tetraploid. At this point, polyploidy doesn't provide evidence for or against junk DNA. However, over time some of the functional DNA of the tetraploid can mutate and become functionless or even be deleted entirely because the other, homoeologous, copy of that functional gene is sufficient to sustain life. This result of polyploidy does provide evidence for junk DNA, and I think it's really an example of #5 above.
One can erect an infinite amount of ad-hoc hypotheses to explain the observed. What matters is what one can evidentially support.
ReplyDeleteI think the first thing you should know if you want to engage in a legitimate scientific discussion about genomic DNA is that genomic DNA sequences can have informational and non-informational roles. That’s Biochemistry 101.
ReplyDeleteDisregarding the non-informational roles of genomic DNA sequences is not only poor science but also disingenuous.
And yet the onion test still stands - why do closely related morphologically similar species need vastly different amounts of DNA playing a non-informational role?
DeleteI don’t know if you are aware, but the person who introduced the ‘onion test’ as a metaphor for the C-value enigma is Ryan Gregory, who throughout his scientific career has developed and promoted Michael Bennett’s ‘nucleotypic hypothesis’ on the biological roles of the so called ‘junk DNA’ (jDNA).
DeleteAccording to Ryan, the ‘nucleotypic hypothesis’, as well as the ‘nucleo-skeletal hypothesis’ proposed by Thomas Cavalier-Smith, “describe genome size variation as the outcome of selection via intermediate of cell size” (emphasis added) (see Ref. 1). You might want to ask Ryan how does he reconcile the ‘nucleotypic hypothesis’ with the ‘onion test’?
In context of my hypothesis on the putative biological role of jDNA as a genomic adaptive defense mechanism against insertion mutagenesis (2), the amount of jDNA varies from one species to another based on the insertional mutagenesis activity and evolutionary constraints on genome size, which explains the overall evolution of genome size (in both single-cell and multicellular organisms) and the c-value enigma (i.e. ‘onion test’).
Considering that in humans and many other species most jDNA originated from viral inserting elements, this genomic defensive mechanism is a classic case of ‘fighting fire with fire’, which is similar to the CRISPR system in which viral sequences have been co-opted in the genome as an adaptive antiviral defense mechanism (3).
According to my evolutionary model, the genomic defense mechanism provided by jDNA is particularly significant against insertional oncogenic transformation by endogenous and exogenous retroviral elements (2). Given the enormous number of somatic cells and high turnover rate during reproductive span of many species such as humans, the number of insertion events, especially by exogenous retroviruses, that would potentially lead to neoplastic transformations in the absence of strong protective mechanisms could be evolutionarily drowning. A dramatic example of the tremendous selection pressure imposed by cancer-inducing insertion mutagenesis in humans is provided by the highly promising biomedical field of gene therapy using viral vectors, a field that has been devastated by high prevalence of cancer in treated patients (4).
1. Gregory TR. Insertion-deletion biases and the evolution of genome size. Gene 2004 Jan 7;324:15-34.:15-34. (http://www.ncbi.nlm.nih.gov/pubmed/14693368)
2. Bandea CI. A protective function for noncoding, or secondary DNA. Med Hypotheses 1990 Jan;31(1):33-4.
3. Horvath P, Barrangou R. CRISPR/Cas, the immune system of bacteria and archaea. Science 2010 Jan 8;327(5962):167-70.
4. Baum C. Insertional mutagenesis in gene therapy and stem cell biology. Curr Opin Hematol 2007 Jul;14(4):337-42.
There seem to be only two possibilities. Either you are misrepresenting Ryan or he's an idiot.
DeleteRyan is a friend of mine. He is not an idiot.
Whether Ryan is a friend of yours, or not, should have no relevance for this discussion. However, Ryan is definitely one of the top scholars in the field of genome evolution, so anyone who wants to "engage in a legitimate scientific discussion about the amount of junk DNA in a genome" should study his papers.
DeleteAs I said, throughout his scientific career, Ryan has developed and promoted Michael Bennett’s nucleotypic hypothesis on the biological roles of the so called ‘junk DNA’ (jDNA). For example, in his first published paper annotated in PubMed (“The Modulation of DNA Content: Proximate Causes and Ultimate Consequences”), which he co-authored with Paul Hebert, they says:
“Although some researchers continue to characterize much variation in genome size as a mere by-product of an intragenomic selfish DNA "free-for-all" there is increasing evidence for the primacy of selection in molding genome sizes via impacts on cell size and division rates” (emphasis added). This paper is available for free at: http://www.ncbi.nlm.nih.gov/pubmed/10207154 ).
Since then, Ryan has published more than a dozen or so of outstanding papers on the evolution of genome size and the nucleotypic hypothesis, although I have yet to find a scientific paper that he published on the so called ‘onion test’.
I think it might be relevant to mention that in his infamous PNAS paper (“Is junk DNA bunk? A critique of ENCODE”; http://www.ncbi.nlm.nih.gov/pubmed/23479647) (see my comment below), Ford Doolittle states that Ryan is “now the principal C-value theorist”, and it appears that he was influenced by Ryan’s work when concluded that:
by developing a “larger theoretical framework, embracing informational and structural roles for DNA, neutral as well as adaptive causes of complexity, and selection as a multilevel phenomenon …much that we now call junk could then become functional.” (emphasis added).
BTW, after studying his papers (or at least reading the 2 quotes above) you might want to reconsider your comment in which you imply that I might have misrepresented his work;
Just to be clear, the 2 quotes from Ryan Gregory’s papers that I was referring to in my previous comment are:
DeleteQuote #1 from Gregory TR, Hebert PD. 1999. The modulation of DNA content: proximate causes and ultimate consequences. Genome Res; 9(4):317-24; http://www.ncbi.nlm.nih.gov/pubmed/10207154):
Although some researchers continue to characterize much variation in genome size as a mere by-product of an intragenomic selfish DNA "free-for-all" there is increasing evidence for the primacy of selection in molding genome sizes via impacts on cell size and division rates (emphasis added).
Quote #2 from Gregory TR. 2004. Insertion-deletion biases and the evolution of genome size. Gene, 324:15-34; http://www.ncbi.nlm.nih.gov/pubmed/14693368):
These are the “nucleoskeletal” and “nucleotypic” theories which, though differing substantially in their specifics, both describe genome size variation as the outcome of selection via the intermediate of cell size (emphasis added)
Although, I do not expect our host Laurence A. Moran to address these quotes, as he usually plays the mute card when confronted with issues that question his believes, hopefully, some of the readers interested in the theories on evolution of genome size would be interested. Of course, one would expect that Ryan Gregory, who is a frequent contributor on this blog, would intervene and set things straight. After all, his ‘friend’ Lary Moran put him in a rather awkward situation: “There seem to be only two possibilities. Either you are misrepresenting Ryan or he's an idiot”. However, that’s also unlikely to happen as I have previously ask Ryan for clarifications on this significant issue, both here and at his site (http://www.genomicron.evolverzone.com/2013/04/critiques-of-encode-in-peer-reviewed-journals/comment-page-1/#comment-8009) and, for whatever reason, he chose to remain silent.
As some of the readers might not have access to Ford Doolittle’s recent PNAS paper (Doolittle WF. 2013. Is junk DNA bunk? A critique of ENCODE”; Proc Natl Acad Sci U S A. 110:5294-300; http://www.ncbi.nlm.nih.gov/pubmed/23479647), it might be relevant to post the full quote in which he addresses the significance of Ryan Gregory’s ‘nucleotypic’ theory:
DeleteOf course, DNA inevitably does have a basic structural role to play, unlinked to specific biochemical activities or the encoding of information relevant to genes and their expression. Centromeres and telomeres exemplify noncoding chromosomal components with specific functions. More generally, DNA as a macromolecule bulks up and gives shape to chromosomes and thus, as many studies show, determines important nuclear and cellular parameters such as division time and size, themselves coupled to organismal development (11⇓–13, 17). The “selfish DNA” scenarios of 1980 (20⇓–22), in which C-value represents only the outcome of conflicts between upward pressure from reproductively competing TEs and downward-directed energetic restraints, have thus, in subsequent decades, yielded to more nuanced understandings. Cavalier-Smith (13, 20) called DNA’s structural and cell biological roles “nucleoskeletal,” considering C-value to be optimized by organism-level natural selection (13, 20). Gregory, now the principal C-value theorist, embraces a more “pluralistic, hierarchical approach” to what he calls “nucleotypic” function (11, 12, 17). A balance between organism-level selection on nuclear structure and cell size, cell division times and developmental rate, selfish genome-level selection favoring replicative expansion, and (as discussed below) supraorganismal (clade-level) selective processes—as well as drift—must all be taken into account.(emphasis added)
Georgi Marinov: And yet the onion test still stands - why do closely related morphologically similar species need vastly different amounts of DNA playing a non-informational role?
DeleteGeorgi,
You asked the question about the ‘onion test’ and I brought up the fact its author, Ryan Gregory, who throughout his career has developed and promoted the ‘nucleotypic’ theory, wrote that:
“Although some researchers continue to characterize much variation in genome size as a mere by-product of an intragenomic selfish DNA "free-for-all" there is increasing evidence for the primacy of selection in molding genome sizes via impacts on cell size and division rates” (1).
“These are the “nucleoskeletal” and “nucleotypic” theories which, though differing substantially in their specifics, both describe genome size variation as the outcome of selection via the intermediate of cell size” (2).
Laurece Moran said that: “There seem to be only two possibilities. Either you are misrepresenting Ryan or he's an idiot.”
How do you interpret the 2 quotes above and the nucleotypic theory?
References
1. Gregory TR, Hebert PD. 1999. The modulation of DNA content: proximate causes and ultimate consequences. Genome Res; 9(4):317-24; http://www.ncbi.nlm.nih.gov/pubmed/10207154
2. Gregory TR. 2004. Insertion-deletion biases and the evolution of genome size. Gene, 324:15-34; http://www.ncbi.nlm.nih.gov/pubmed/14693368
Someone definitely doesn't know those things:
ReplyDeletehttp://www.worldscientific.com/doi/pdf/10.1142/9789814508728_0009
Oh, and I forgot to mention that it contains sentences like the following:
DeleteNot only is some RNA transcribed from the antisense strand, but RNAs can also be transcribed from several different start
sites within an open reading frame. So a single open reading frame can carry multiple overlapping codes that specify both protein-coding RNAs and non-protein-coding RNAs
Georgi, when you give such a link, you should at least tell us what it is before we click on it.
DeleteIt's Jonathan Wells 2013, from the IDiot conference proceedings "Biological Information", showing he has learned nothing in 20 years and still lies about DNA. Not just junk.
Consider the first four sentences from the Abstract of the Wells toilet paper that Georgi linked to. All of his first four sentences demonstrate profound, mind-boggling ignorace of genetics and scientific language.
I'll number the sentences.
[Jonathan Wells, 2013]: [1] In the 1950s Francis Crick formulated the Central Dogma of molecular biology, which states (in effect) that DNA makes RNA makes protein makes us.
[2] By 1970, however, biologists knew that the vast majority of our genome does not encode proteins, and the non-protein-coding fraction became known as “ junk DNA.”
[3] Yet data from recent genome projects show that most nuclear DNA is transcribed into RNAs, many of which perform important functions in cells and tissues.
[4] Like protein-coding DNA, non-protein-coding regions carry multiple overlapping codes that profoundly affect gene expression and other cellular processes.
Everyone here knows what's wrong with that, but for the hell of it:
[1] Wells doesn't know what the Central Dogma is. It is not "DNA makes RNA makes protein makes us." Larry has posted on this many times. Wells knew nothing 20 years ago and has learned nothing.
[2] "the non-protein-coding fraction became known as “ junk DNA.”" Lie. No geneticist or molecular biologist ever said that non-coding DNA was equal to junk DNA or a subset of it. No ID creationist including Wells or Luskin has ever produced a single quote from a geneticist or molecular biologist ever saying that he himself believed non-coding DNA was equal to junk DNA.
[3]. "RNAs, many of which perform important functions". This is "Much-munching": lying by the using words like "much" or "many" which in fact refer to a tiny, tiny percentage of the genome, less than 10 percent.
[4] "non-protein-coding regions carry multiple overlapping codes" This IDiot just said there are multiple codes in DNA. NO. DNA has one code, many sequences. What's even stupider [if that's even POSSIBLE!] is that this IDiot just said non-protein-coding regions code! How can something NON-CODING have a CODE!? And SEQUENCE is not a CODE! There's ONLY ONE CODE per GENOME!
Ignoramus. We've corrected them thousands of times, they're still dumb as dishwater.
1. You seem to be protesting too much. That was pretty much the way Crick presented the Central Dogma when he coined that unfortunate label. I assume Wells added the "makes us" because his audience has no idea what protein is.
Delete2. It's an exaggeration. I cannot tell how severe without more context. If Wells was trying to turn around and show that these biologists are liars because some non-coding DNA serves a clear purpose, then yes, he is a liar.
3. Yes, but 10% is not "tiny tiny", either. I would have to call what you did "much-munching."
4. A little confusing. Is your complaint simply that he used the wrong terms? Isn't there something more substantial there to justify all your shouting?
John Vreeland says,
Delete1. You seem to be protesting too much. That was pretty much the way Crick presented the Central Dogma when he coined that unfortunate label.
Diogenes alerted you to the possibility that your view might be incorrect so why didn't you do a bit of research before revealing your ignorance?
That is NOT what Francis Crick said when he published his original paper and it's not what he said when he published the follow-up paper in Nature in 1970. Please read Basic Concepts: The Central Dogma of Molecular Biology.
You may want to post another comment after you have informed yourself of the facts.
2. It's an exaggeration. I cannot tell how severe without more context. If Wells was trying to turn around and show that these biologists are liars because some non-coding DNA serves a clear purpose, then yes, he is a liar.
There's not much doubt that Wells is a liar. You can read the documentation at Junk & Jonathan: Part 1—Getting the History Correct. If you want to read the entire series try: The Myth of Junk DNA by Jonathan Wells
I'm happy to discuss this with you once you have read all my previous posts.
Someone needs to coin another word besides "junk". Junk implies it is worthless. Maybe it was worthwhile in the past but it's out there at the dumpster ready to be taken away by the trash collector because it has little future use.
ReplyDeleteBut that really does not describe "junk" DNA. Introns and gene duplication allowed evolution a lot of elbow room to quickly make variations of genes with alternative splicing as just one example. Those trash pseudogenes may have played an important part in the evolutionary scheme in the past or may become important in the future.
Call them "not quite junk" DNA or "check back later when you are fighting off some new disease you might like me more" DNA, or "haven't figured out the epigenetics yet here" DNA.
Just because you can mutate the hell out of it, or delete it completely and still make it to tea does not mean it's true "junk" DNA.
Some years ago I noticed that there were two kinds of rubbish in the world and that most languages have different words to distinguish them. There is the rubbish we keep, which is junk, and the rubbish we throw away, which is garbage. The excess DNA in our genomes is junk, and is there because it is harmless, as well as being useless, and because the molecular processes generating extra DNA outpace those getting rid of it. Were the extra DNA to become disadvantageous, it would become subject to selection, just as junk that takes up too much space, or is beginning to smell, is instantly converted garbage. - Sydney Brenner
Deletehttp://sandwalk.blogspot.ca/2012/09/stephan-jay-gould-and-sydney-brenner.html
Dan Graur will talk about functional, indifferent, junk, garbage, and zombie DNA on SMBE 2013. See abstract 283 in this pdf
DeletePaul Morrison says,
DeleteJust because you can mutate the hell out of it, or delete it completely and still make it to tea does not mean it's true "junk" DNA.
This is correct. However, the known facts greatly limit the possibilities and most proponents of function fail to recognize that.
BTW, you can't argue that the reason we have lots of junk in our genome is in order to prepare for the future. Evolution doesn't work that way. You're confusing it with Intelligent Design Creationism.
"BTW, you can't argue that the reason we have lots of junk in our genome is in order to prepare for the future. Evolution doesn't work that way. You're confusing it with Intelligent Design Creationism."
DeleteI think it is possible to envision a crude mechanism in which this can happen in a non-teleological way. Imagine two lineages, one with little junk DNA and one with a lot. Since junk DNA can serve as a source of new variation, the lineage with a lot of junk may tend to have an higher probability of adapting to new conditions than the other one simply because it has a bigger source of DNA evolving neutrally that can randomly provide new solutions. Thus on evolutionary time-scales the lineage with more junk can end up surviving better than the other. In this way you could say that the reason is "preparing for the future", although this is a simplification that should not be taken literally (although IDiots would certainly take it so and present it out of context).
Obviously, this leaves out the question of why there are so many close lineages with so much differing junk DNA.
Pedro,
DeleteYou make a good point, but because it goes against the prevalent view it is going to be ignored (pretending that an idea does not exist is much more efficient in keeping it silent than by criticizing it!).
In one of my comments bellow, I explained why so many close lineages have such high differences in their c-values: it has to do with the activity of viral and transposable elements, which can vary not only between distant lineages but also highly related lineages. See, for example, the high differences in the Alu transposition activity between human and chimpanzee lineages (1).
1.Hedges DJ et al. 2004. Differential alu mobilization and polymorphism among the human and chimpanzee lineages. Genome Res. 14(6):1068-75. (available for free at: http://www.ncbi.nlm.nih.gov/pubmed/15173113)
Pedro's idea is interesting, but my understanding was that a lot of "junk" DNA was simply multiple repeats of the same simple ... junk. I don't see it as having the same richness of utility that a simple gene duplication event has.
DeleteAlso, it is important to emphasize that evolution cannot "pre-adapt" anything. Extra unused DNA might (maybe?) provide a useful source for new genes, but until it is actually selected its contents will be constantly mutating throughout the history of the species. "Pre-genes" cannot lie dormant for eons waiting to be awakened by events. All useful features must be preserved by natural selection, which means you have to use it or you will lose it. One of the ID gang seemed confused on that point a while back, which usually means they will continue to be confused about it.
" I don't see it as having the same richness of utility that a simple gene duplication event has."
DeleteTrue, but pseudogenes are junk DNA too, but it's perfectly conceivable IMHO that a small part may end up producing something useful later on. Even parts of these pseudo-genes may end up being useful. I'm not the right person to ask details, though.
"Also, it is important to emphasize that evolution cannot "pre-adapt" anything."
Certainly, I agree. But the crude idea I mentioned above has nothing to do with pre-adaptation.
And yes, the whole ideia that functional genes can remain perfectly functional sitting there doing nothing for millions of years is one of the main problems with concepts like Cosmic Ancestry (a flavour of Panspermia highly inspired by Hoyle&Wickramasinghe) or with the nonsense that Paul Davies came up with as an explanation for cancer mechanisms. These people haven't realized yet that if a gene sequence is unused than nothing will stop it (i.e. NS) from accumulating mutations and pseudogenization will follow unavoidably. This isn't even speculation, it's one of the most solid concepts we have, fully supported by evidence.
The conceptual, evolutionary and experimental evidence that most of the genomic DNA in humans and many other species with high C-values does not have informational roles has been overwhelming for more than three decades. For this reason, many scholars in this field have focused on potential non-informational roles for most of the genome. See for example the ‘nucleo-skeletal hypothesis’ proposed by Thomas Cavalier-Smith and the ‘nucleotypic hypothesis’ proposed by Michael Bennett and Ryan Gregory.
ReplyDeleteTherefore, in light of this evidence and scholarly work, focusing on, or arguing that most of the so called ‘junk DNA’ (jDNA) *does* or *does not* have informational roles is not only poor science but disingenuous.
A rather surprising example of this disingenuity is displayed in the recent PNAS paper by Ford Doolittle (“Is junk DNA bunk? A critique of ENCODE”; http://www.ncbi.nlm.nih.gov/pubmed/23479647) in which he states that ‘junk DNA’ has been traditionally referred to, or understood, as:
“DNA that does not encode information promoting the survival and reproduction of the organisms that bear it” (emphasis added), and that
“I submit that, up until now, junk has been used to denote DNA whose presence cannot reasonably be explained by natural selection at the level of the organism for encoded informational roles” (emphasis added).
However, as mentioned above, ever since the concept of jDNA was introduced, the primary putative roles attributed to jDNA as a group of genomic sequences were non-informational roles (e.g. nucleo-skeletal and nucleotypic hypotheses) and to imply otherwise is misleading. Apparently Ford Doolittle built his ‘strawman’ to be able to take it down in his conclusion:
by developing a “larger theoretical framework, embracing informational and structural roles for DNA, neutral as well as adaptive causes of complexity, and selection as a multilevel phenomenon …much that we now call junk could then become functional.” (emphasis added).
I'm quite well aware of the possibility that the extra DNA might have a role that doesn't depend on its sequences. My goal in this post is to make sure everyone knows the facts so they know what possibilities are still viable.
DeleteIt is no longer reasonable to postulate that most of our genome has a function that depends on its sequence. While there might be other possibilities, none of them seem very reasonable at this time and none of them pass the onion test.
Ever since the C-value paradox was defined half of century ago and it has become evident that most genomic DNA in complex organisms is composed viral and transposable elements, it was no “longer reasonable to postulate that most of our genome has a function that depends on its sequence”. That’s why most of the scholars in this field (e.g. Michael Bennett, Thomas Cavalier-Smith, Dmitri Petrov, Ryan Gregory) have focused on potential non-informational roles.
DeleteTherefore, it would make sense that the discussion should focus on potential non-informational roles for the bulk of the genome and, I discussed in my comment above and many other comments here at Sandwalk and elsewhere, some of these putative roles (e.g. defense mechanism against insertion mutagenesis, particularly against insertional oncogenic transformation) might explain the evolution of genome size and the C-value enigma, including the ‘onion test’.
In context of the model I proposed, as a genomic adaptive defense mechanism against insertion mutagenesis, the amount of the so called ‘junk DNA’ varies from one species to another based on the insertional mutagenesis activity and evolutionary constraints on genome size.
In one of my comments above (Saturday, July 06, 2013 1:41:00 PM), I mentioned that, similar to the CRISPR/Cas protective system, the genomic defensive mechanism against insertion mutagenesis provided by the so called ‘junk DNA’ (jDNA) is a classic case of fighting fire with fire. Interestingly, similar to jDNA, the presence and the quantity of viral DNA sequences co-opted by the host in the adaptive CRISPR/Cas antiviral defense mechanism varies from one host to another depending on the viral activity and on the constrains on the genome size.
DeleteIn context of my hypothesis on the putative biological role of jDNA as a genomic adaptive defense mechanism against insertion mutagenesis (2), the amount of jDNA varies from one species to another based on the insertional mutagenesis activity and evolutionary constraints on genome size, which explains the overall evolution of genome size (in both single-cell and multicellular organisms) and the c-value enigma (i.e. ‘onion test’).
ReplyDelete"[A]gainst insertion mutagenesis...." - Looks to me an awful lot like Larry's #1.
Re using this as some actual non-sequence-dependent functional role that explains the Onion test," please provide references for the extreme variation in mutational proclivity in the genomes of closely related species that would be necessary to explain the extreme variations in genome size.
My model on the biological role of the so called ‘junk DNA’ (jDNA) emphasize it protective function against insertional oncogenic transformation of somatic cells, which have an extraordinary turn over during the life span of complex organisms, such as humans, and are exposed not only to insertion mutagenesis by endogenous mobile elements but also by exogenous viral elements, such as retroviruses, whose genome must insert into the host cells genome in order lead to a productive life cycle.
DeleteIt is well known that most of the genomic jDNA sequences represent endogenous viral and transposable elements and their remnants. Therefore, the quantity of jDNA in the genome of various species, including closely related species, is based on the difference in the activity of these endogenous elements, as well as those of the exogenous inserting elements, in the germ line. This insertional activity can vary greatly not only between distantly related species but also between closely related species, such as onions, which explains the difference in the quantities of their jDNA sequences. This fact is supported by the entire scientific literature on genome composition of various species.
As a classic example of ‘fighting fire with fire’, the amount of jDNA sequences (and protection) increases with the increase in insertion activity and vice versa. However, the quantity of jDNA that accumulates in the genome is also influenced by differential constrains (e.g. energetic costs) on genome size among various species, such as humans and hummingbirds for example. Nevertheless, these constrains are usually similar in related species, such as onions, so the difference in these species is based primarily on the rate of insertional activity.
"Therefore, the quantity of jDNA in the genome of various species, including closely related species, is based on the difference in the activity of these endogenous elements, as well as those of the exogenous inserting elements, in the germ line. This insertional activity can vary greatly not only between distantly related species but also between closely related species, such as onions, which explains the difference in the quantities of their jDNA sequences."
DeleteCorrect me if I'm wrong, but this sounds like circular reasoning to me. Insertional activity is seen by the amount of insertion sequences in DNA, so the explanation of higher quantities of junk and insertional sequences in close related species is because they have different insertional activity? That they have different insertional activity and amounts of junk is what we observe, but that does not explain why that is the case. You're explaining the observation with the observation itself, or so it seems to me.
In answer to my request for references to support the extreme variations in mutational load between closely related species that would be needed to establish the c-value paradox as the sort of genetic defense mechanism he proposes, Claudiu replies:
DeleteThis fact is supported by the entire scientific literature on genome composition of various species.
For some reason this reminds me of the classic Sarah Palin response to Katie Couric when asked which newspapers she read regularly:
"All of them."
@Pedro,
DeleteIn various species, including closely related species (in which it is expected that the rate of sequence deletion is similar), the quantity of jDNA sequences that accumulate in the genome depends on the activity of viral and transposable elements. Based on this activity, even highly related species can accumulate various amounts of jDNA sequences within a relatively short period of evolution.
As I previously mentioned, similar the quantity of viral DNA sequences co-opted by some of bacterial and archaeal species in their adaptive CRISPR/Cas antiviral defense mechanism, which depends on the viral activity, the amount of protective jDNA sequences that accumulate in the genome of various species depends on the activity of viral and transposable elements.
@judmarc,
DeleteMost of the jDNA sequences in the genome of various species originated from the activity of viral and transposable elements. This fact is fully supported by all studies on genomes composition performed in the last half of century. You might want to read some of them.
As pedro mentioned, there is a difference between a fact and an explanation. You haven't yet explained why there should be such extreme differences between closely related species regarding activity of viral and transposable elements. As pedro says, you keep trying to use the fact (the difference in the amount of junk DNA) as the explanation for itself: A difference in the amount of junk DNA = a difference in the amount of viral and transposable element activity, which is "proved" by a difference in the amount of junk DNA, and back we are at the beginning of the circle.
DeleteSeems to me you're seeking adaptationist reasons where simple randomness (e.g., polyploidy) would suffice.
Interesting that you brought up ‘polyploidy’, which Barbara discussed in her comment above. Polyploidization is a fact. Isn’t this ‘fact’ the ‘explanation’ for the he large difference in genome size (not c-value) between some highly related lineages?
DeleteIsn’t this ‘fact’ the ‘explanation’ for the he large difference in genome size (not c-value) between some highly related lineages?
DeleteAnd since it is a random process that we know works this way because we see it happen today, we have a reasonably complete explanation.
You, on the other hand, propose non-random processes (e.g., one genome being far more susceptible to insertions of transposable elements than another closely related genome), with no known mechanism of action, that we don't see happening today. Thus your "explanation" is missing at least a couple of steps.
judmark,
DeleteI presume that you didn’t have the chance to read one of my earlier comments today:
In one of my comments bellow, I explained why so many close lineages have such high differences in their c-values: it has to do with the activity of viral and transposable elements, which can vary not only between distant lineages but also highly related lineages. See, for example, the high differences in the Alu transposition activity between human and chimpanzee lineages (1).
1.Hedges DJ et al. 2004. Differential alu mobilization and polymorphism among the human and chimpanzee lineages. Genome Res. 14(6):1068-75. (available for free at: http://www.ncbi.nlm.nih.gov/pubmed/15173113)
So, to paraphrase your comment “since it (i.e. production of jDNA) is a random process that we know works this way because we see it happen today, we have a reasonably complete explanation” for differential accumulation of jDNA in highly related lineages.
Claudiu,
DeleteNo.
First, the article is about a twofold increase in Alu insertions in humans vs. chimpanzees, not a manyfold increase in size of the entire genome, as with the onion paradox.
Second, this: "Their results suggest that substantial variation in transposition and/or fixation rates may exist among primate lineages. Whether these differences are attributable to underlying differences in biology, stochastic fluctuations in Alu proliferation, and/or broader population–genetic processes remains to be determined." [Emphasis added.]
Reading further into the article, this is an outline of the authors' present hypothesis:
DeleteThe basic components of our model are as follows: (1) variation in source Alu-producing loci exist in the population, (2) stochastic sampling of these source variants either at speciation or during bottleneck events alters the population-level Alu transposition activity (insertions per birth), and (3) although the previous two conditions are sufficient to produce variation within and between lineages, smaller effective population sizes will both increase the sampling variance of Alu sources and reduce a given population's ability to select against deleterious source loci. This may result in a substantially increased population-level Alu activity (insertions per birth) brought about by environmental insults, speciation events, etc.
This is exactly the opposite of what we should expect if your hypothesis is true and junk DNA such as Alu elements represents a positive adaptation to genetic load, rather than a nearly neutral or slightly deleterious mutation that must be accounted for by random sampling events amplified by small effective population sizes.
judmark: This is exactly the opposite of what we should expect if your hypothesis is true and junk DNA such as Alu elements represents a positive adaptation to genetic load…
DeleteMy hypothesis dos not imply that the so called ‘junk’ DNA sequences, such as Alu elements, represent a positive adaptation to genetic load. You might want to take a look at these posts for on outline of the model:
http://sandwalk.blogspot.com/2012/06/tributre-to-stephen-jay-gould.html
http://comments.sciencemag.org/content/10.1126/science.337.6099.1159
http://thefinchandpea.com/2012/09/13/my-last-thoughts-on-the-media-coverage-of-encode/#comments
My model on the biological role of the so called ‘junk DNA’ (jDNA) emphasize it protective function against insertional oncogenic transformation of somatic cells
DeleteMy hypothesis dos not imply that the so called ‘junk’ DNA sequences, such as Alu elements, represent a positive adaptation to genetic load.
OK, now I'm confused.
http://evolutionwiki.org/wiki/Genetic_load
DeleteSo these virii and transposons do insertional oncogenic transformation, but hold the line at other harmful recessive mutations? Why, out of a sense of morals?
Deletejudmarc: So these virii and transposons do insertional oncogenic transformation, but hold the line at other harmful recessive mutations?
DeleteNo, they don’t. However, I think this discussion has deviated from the essence of the my hypothesis that, in some species such as humans, the so called ‘junk DNA’ (jDNA) which occupies more than 90% of the human genome, provides significant protection against deleterious insertional mutagenesis by endogenous and exogenous inserting elements, particularly against insertional oncogenic transformation.
When it comes to the ‘onion test’, we should all ask Ryan Gregory how does he reconcile the nucleotypic hypothesis, which he has developed and promoted throughout his scientific career, with the test (for references and quotes on nucleotypic hypothesis see my other comment on this posts).
BTW, what do you think about nucleotypic hypothesis concerning the biological function of jDNA, which has been presented and discussed in dozens of publications?
I read a couple of Dr. Ryan's publications. They seem to me to be good, careful work. Although I am not nearly qualified to assess the part of Dr. Moran's comment where he says of Dr. Ryan "or he's an idiot," to the extent I as an amateur am able to evaluate, I do agree with Dr. Moran about the first part of his comment, i.e., that you appear to be misrepresenting Dr. Ryan's work.
DeleteI would have included the possibility you were misunderstanding his work, but the clarity of that work, combined with your intentional tactics such as careful partial quoting to create a misleading impression, make me believe Dr. Moran is almost certainly correct to call it misrepresentation.
judmarc: I read a couple of Dr. Ryan's publications. They seem to me to be good, careful work
DeleteAs I wrote in previous comments on this post, Ryan’s work is more than ‘good’ and ‘careful’, it is outstanding:
Ryan has published more than a dozen or so of outstanding papers on the evolution of genome size and the nucleotypic hypothesis… Ryan is definitely one of the top scholars in the field of genome evolution, so anyone who wants to "engage in a legitimate scientific discussion about the amount of junk DNA in a genome" should study his papers.
The problem, however, is many people who are participating in the junk DNA debate have not studied his work, or they don’t understand it.
Apparently you read Ryan’s papers, so what do you think about his nucleotypic hypothesis?
judmarc: I do agree with Dr. Moran about the first part of his comment, i.e., that you appear to be misrepresenting Dr. Ryan's work
DeleteWhat’s to misrepresent about the following statements:
Although some researchers continue to characterize much variation in genome size as a mere by-product of an intragenomic selfish DNA "free-for-all" there is increasing evidence for the primacy of selection in molding genome sizes via impacts on cell size and division rates (emphasis added) (Gregory TR, Hebert PD. 1999. The modulation of DNA content: proximate causes and ultimate consequences. Genome Res; 9(4):317-24).
“These are the “nucleoskeletal” and “nucleotypic” theories which, though differing substantially in their specifics, both describe genome size variation as the outcome of selection via the intermediate of cell size” (2). (emphasis added) (Gregory TR. 2004. Insertion-deletion biases and the evolution of genome size. Gene, 324:15-34).
What confusion or misrepresentation is here? These quotes represent the essence of the nucleotypic hypothesis. Again, what do you think that the nucleotypic hypothesis is?
If there are about 100 mutations per human child, wouldn't that mean in only 10,000 years there would be a million mutations? But 10,000 years ago, human beings were essentially the same as they are now? Were these million mutations all junk dna?
ReplyDeleteBut 10,000 years ago, human beings were essentially the same as they are now? Were these million mutations all [in] junk dna?
DeleteMost of them would be of course. But even in definitely non-junk DNA, protein coding genes for example, many (perhaps most) mutations will be neutral in effect or nearly so. As for seriously deleterious mutations, they will have went with their carriers into the ground.
Anyway, we cannot possibly be the same as our ancestors on the molecular level in the first place, as change over time is inevitable. One might well detect genetics-based phenotypic differences over 10,000 years but even if not, that doesn't mean the nucleotide sequence has remained static. Also, take note that 1 million nt represents only about 0.03% of the genome.
The mutation rate is 100 mutations PER GENERATION. In 10,000 years there would be about 350 generations and 35,000 mutations.
DeleteI missed that error.
DeleteWell, if humans and chimpanzees had a common ancestor 7 million years ago, that would be 24500000 mutations(7*(35,000*100)) .
DeleteEven if 98% of the mutations were from junk dna, the other 2%(490,000) would from protein coding dna. Now, if the haploid human genome contains approximately 20,000 protein-coding genes(which according to the human genome project, it does), that would mean that every protein coding gene would have gotten mutated during those 7 million years, and many would've been mutated many times! Yet the phenotypical differences between chimps and humans are very similar.
Now, I understand biology isn't always mathematically linear, but this margin of error is just seems ridiculous.
I reckon there would have been less mutations in the past, since we're exposed to many artificial mutagens now.
@Johnny Diala,
DeleteMany of the mutations in the protein coding genes would have been deleterious and removed by natural selection, i.e. the animal with the deleterious mutation would not have had any offspring and not passed that mutation along.
And many of the mutations, even in protein coding genes, are silent and thus at most minimally visible to selection. As it happens, the average difference between humans and chimps within protein-coding exons is 0.5%, or about one per 200 bases. If we suppose the average protein to be about 300 amino acids, that's around 5 differences per protein. Is that enough to give you "many times"? And of course the approximately three-fold difference in distance between exons and junk suggests that around 2/3 of the mutations in exons were deleterious, which is in fact just about the proportion of non-silent sites.
DeleteJohnny has miscalculated the expected differences: he's estimateed the number in the human lineage, but there should be an equal number in the chimp lineage. And in fact there are about 40 million differences.
Whoops. Larry, I think you are misleading Johhny a bit here. He confuses the number of mutations with the number of fixations, and so do you. I presume that isn't what you meant to say. There should be 100 mutations per generation per individual(or per haploid genome?; never mind, doesn't matter here) and, through the magic of neutral evolution, 100 fixations per generation in the population. But there would be millions of mutations per generation in the population, only 1/(2N) of which would ever be fixed. Your point about generations instead of years remains valid, though.
DeleteJohn says: "the average difference between humans and chimps within protein-coding exons is 0.5%, or about one per 200 bases."
DeleteFascinating-- did you get that from the chimp genome paper, or somewhere else?
"But there would be millions of mutations per generation in the population, only 1/(2N) of which would ever be fixed."
Is it really 1/2N? I thought the probability of an individual neutral mutation getting fixed was 1/N, if N = population size.
For the record, most proteins differ by just one or two amino acids, if compared between human and chimp. Many are identical in terms of amino acids, not at the genetic level.
Delete@John Harshman,
DeleteThanks for clarifying.
Fir a discussion of the number of differences between humans and chimps and how it relates to the mutation rate, see ...
DeleteEstimating the Human Mutation Rate: Phylogenetic Method
Is it really 1/2N? I thought the probability of an individual neutral mutation getting fixed was 1/N, if N = population size.
DeleteThat would be true for a haploid population, but not for a diploid one. Or, to put it another way, if N refers to the population of human individuals, it's 1/(2N); but if N refers to the population of alleles, it's 1/N. The former is customary.
@Steve
DeleteThe mutations would not have necessarily been deleterious enough to be selected against. If you assume 100 mutations per generation, and then you go back from your ancestory(grandfather, great-grandfather, great-great grandfather), it would still be the same number of mutations(since each of your ancestors got about a 100 mutations), yet none of them would have been selected against since if they were you wouldn't be here.
Oh, and Larry (or Dr. Moran, if you'd prefer), your 350 generations per 10,000 years may be an underestimate. You're assuming one generation is approx 28 years.
DeleteI'm not an anthropologist, but I'm fairly certain that generations wouldn't have been more than 15-16 years apart throughout most of human history. Especially since the life expectancy rarely exceeded 30 or 40.
Johnny, you're still thinking in terms of mutations when you should be thinking about fixations. The probability of a neutral mutation becoming fixed in the population is its current frequency. Given that the frequency begins at 1 divided by the population size (2N in the case of a single mutation in a diploid species), almost all of your grandfather's mutations will eventually become extinct. Most of them have already. Remember that you only inherit half of each parent's genome. Mutations disappear from the population all the time without the need for the deaths of any lineages.
DeleteWell then can't you just divide it by two every generation since you only inherit half of them?
DeleteI'm not looking at the genetics of the population as a whole, rather what happens to those mutations directly within an ancestral lineage. When looking directly at the lineage, I don't think it's necessary to see whether or not it's fixated within the population, but rather see what happens to it directly within a single lineage. Since those mutations wouldn't have been selected against, the only real way they could have been eliminated is by meiosis.
True. So you expect to inherit 100 mutations from each generation (50 from each parent, 25 from each grandparent, and so on.) If the population were infinite, 100 mutations per generation in a lineage would be an adequate description. But the population isn't infinite. Go back enough generations and you get repeat ancestors, e.g your great^10 grandmother on your mother's side is also your great^9 grandmother on your father's side, so the number of mutations per generation goes down. Go back far enough and the number of mutations per generation that are expected to contribute to you goes down to zero, just because of all the repeats. (This is another, fairly odd, way to think about the phenomenon called coalescence.)
DeleteIt would've taken more than 10 generations to get repeat ancestors, especially if you're the product of miscegenation.
DeleteYou'd be surprised. But the numbers don't matter as long as you understand the concept.
DeleteOf course if you receive that same deleterious mutation from two different lines of ancestry then you are removed from the gene pool.
DeleteYou do if it's a recessive lethal. Otherwise, probably not.
DeleteI did say it was deleterious. Okay, perhaps not specific enough.
DeleteYes it does seem incontestable that a large portion of the genome does not carry out any functions by virtue of its base sequences.
ReplyDeleteBut there does seem to be an interesting question about genetic load. Wouldn't we have to measure the effects of genetic load by comparing the variation in mutations in surviving embryos to the variation in mutations in all conceptions? If ninety percent of base pairs are junk, the simple (simplistic I suspect) math is that there would only be odds of 2 700 000 to one that a mutation would have any effect at all? What it really seems like is that genetic load is a determined by effectiveness of replication and genome repair mechanisms, unrelated to any question about junk DNA at all.
I don't understand your math. If 90% is junk, one might naively think that the chance of a mutation having any effect at all is 1/10. (Naively because non-junk contains many neutrally evolving sites and others for which mutations have minimal effect.)
DeleteNinety percent of three billion base pairs is actually 2 700 000, not 2 700 000, I dropped a zero. Ten percent of three billion is then actually 3 000 000. The three billion is a crude estimate of the number of base pairs in the human genome. Using the estimates instead of the percentages raises the question of whether 100 mutations is a heavy genetic load, given the high probability none of the mutations has any effect whatsoever. Why not ten thousand?
DeleteMy question was, is there a difference in the number of mutations in failed embryos than in surviving embryos? (I'm sorry I clumsily phrased this as variation in mutation, instead of variation in mutation rates.) If there is, it would be evidence that changes in junk DNA have some sort of role, albeit no significant one from the base pair sequences. My guess is that the rate of variations in failed embryos is pretty much the same as the rate of variations in the whole population. Genetic load is more about the messiness of chemistry than the neatness of mathematical genetics. I think. The informed may know better, but that is my question.
Your numbers are still way off. Ten percent of three billion is 300,000,000. But I doubt your experiment would show anything. I don't know the variance in numbers of mutations. If it were a poisson process the variance would be about 100, but perhaps it isn't. Now the more mutations, the greater the probability that one of them would be lethal, or that the combination of several would be. But that's true regardless of how much of the genome is junk. Whether you're aiming at a small target or a big one, increasing the number of shots also increases the number of hits. I doubt, however, that any experiment would give a significant result.
Delete"Now the more mutations, the greater the probability that one of them would be lethal, or that the combination of several would be. But that's true regardless of how much of the genome is junk." This is what I can't follow. If 2 700 000 000 base pairs are junk, the hundred mutations have very low odds effecting a meaningful change, namely, 2 700 000 000 to one. Now it's true that doubling the mutation rate increases the odds of having a lethal effect necessarily increase. But if essentially all the DNA has some role, then the odds that a mutation will do something, good, bad or neutral, are essentially one out of one. Then, a variation in mutation rate in viable and nonviable embryos is a measure of the significance of the ninety percent's unknown function.
ReplyDeleteYes, it's true that increasing the mutation rate increases the chances of lethality, just as buying a hundred lottery tickets increases the chances of winning the grand prize, by some minuscule amount. But all DNA having a function is analogous to increasing the number of winning combinations, something much more significant given the absolute numbers. (Aside from my tendency to mistype the numbers, the odds of neutral and positive mutations is pretty much ignored, aside from noting that the question is whether there's a difference in genetic load in viable and nonviable embryos.)
But the question is whether a hundred mutations is a break even point, the number of mutations that generally will not vitiate an embryo? My guess is that this mutation rate is set by biochemical necessity, and that the genetic load of the ninety percent unknown function (aka junk) DNA is not a load in the sense of being static in the genetic signal. As to what checking this would show, it would merely be another, I think independent, confirmation that junk DNA is junk. Also, more generally it would confirm the prevalence of neutral mutations. And it would suggest that studies of genetic variation need to try to filter out the variation in junk DNA, measuring relevant variation in what you might call the core genome.
Nothing earth shaking I grant you, but still interesting. Negative results seem to me to be science too.
I don't think your method of estimating the proportion of junk would work. Too little statistical power. But the number of mutations we do see corresponds well with the number predicted by many methods of estimating mutation rates (see Larry's several articles on the subject) and besides, we know that most of the genome is junk by a more powerful method: by its lack of sequence conservation over evolutionary time.
DeleteJohn,
DeleteThe lack of sequence conservation only supports the paradigm that it doesn’t have an informational function; genomic DNA can play various non-informational roles.
I refuse to feed your obsession. Sorry.
DeleteJohn,
DeleteApparently, you are not up to date with the thinking in this field. The idea that most of the genome in species with high c-value may play informational roles has been abandoned many decades ago by the scholars in the field.
Since then they have been focusing on potential non-informative roles.
Here are some excerpts from Ford Doolittle’s recent PNAS paper on junk DNA and c-value enigma (Doolittle WF. 2013. Is junk DNA bunk? A critique of ENCODE”; Proc Natl Acad Sci U S A. 110:5294-300):
“Cavalier-Smith (13, 20) called DNA’s structural and cell biological roles “nucleoskeletal,” considering C-value to be optimized by organism-level natural selection (13, 20). Gregory, now the principal C-value theorist, embraces a more “pluralistic, hierarchical approach” to what he calls “nucleotypic” function (11, 12, 17). A balance between organism-level selection on nuclear structure and cell size, cell division times and developmental rate, selfish genome-level selection favoring replicative expansion, and (as discussed below) supraorganismal (clade-level) selective processes—as well as drift—must all be taken into account.
He also wrote that: by developing a “larger theoretical framework, embracing informational and structural roles for DNA, neutral as well as adaptive causes of complexity, and selection as a multilevel phenomenon …much that we now call junk could then become functional.” (emphasis added).
BTW John, what do you think about Gregory’s nucleotypic hypothesis?
What part of "I refuse to feed your obsession" is unclear to you?
DeleteThe part in which you erroneously write “we know that most of the genome is junk”, when the experts in the field, including Cavalier-Smith, Gregory and Doolittle, say that might not be the case.
DeleteWhat part of "part of" is unclear to you? The bit you quote wasn't even part of the post you were replying to. And I still refuse to feed your obsession.
DeleteI responding to S Johnson’s comment you misleadingly wrote, with apparent authority, that: “*we* know that most of the genome is junk by a more powerful method: by its lack of sequence conservation over evolutionary time”. That’s not true! Lack of sequence conservation does not prove that a genomic sequence is non-functional, or ‘junk’. For this and other reasons, the scholars in this field, including Thomas Cavalier-Smith, Michael Bennett, Ryan Gregory have focused on non-informational roles for most of the genome in organisms with high-c value.
ReplyDeleteUnfortunately, it appears that you as well as some of the other contributors to this post entitled “Five Things You Should Know if You Want to Participate in the Junk DNA Debate”, including the author of the post, do not know or understand the ‘nucleotypic hypothesis’ proposed by Michael Bennett and Ryan Gregory.
However, that might be somewhat excusable, because very few genome researchers have referred to, or discussed these theories. Instead, some of them postulated various informational roles for the so called ‘junk’ DNA (jDNA) despite overwhelming evidence, such a lack of sequence conservation, that it doesn't have informational roles. I believe this was the main this reason why Ryan came with the simple c-value reality: the ‘onion test’.
The obvious and interesting question is: does Ryan’s nucleotypic hypothesis passes the onion test?
Hi Professor
ReplyDeleteI just have one question about your first point. If each subsequent generation has about 100 mutations not found in either parent and 10% of our genome is functional, then this would imply that on average each person has an additional 10 mutations that affect functional regions of our DNA. Surely our birth rate has never been high enough for this to be sustained over a long period of time.
Here is a post he made, with a bit more information on the topic...
Deletehttp://sandwalk.blogspot.de/2009/11/genetic-load-neutral-theory-and-junk.html
Additionally, some mutations can lead to an organism dying before birth (possibly very well before birth, before the woman even knows she's pregnant).
DeleteThis means there are some humans that we're not even counting, if we're looking at survival from birth or mutations in grownups or something similar. Some of the worst-mutated humans have already been winnowed out well before then.
It's a selection bias effect.