If you want to be a serious participant in the debate over junk DNA then you should watch this video. Dan Graur presents the standard arguments for junk DNA—most of which have been around for decades. He also destroys the main arguments against junk DNA. You are entitled to choose sides in this debate but you are not entitled to pose as an authority unless you know the best arguments from BOTH sides. It is not sufficient to just quote evidence for function as support for your bias. You must also refute the evidence for junk. You have to show why it is wrong or misleading.
Hat Tip: PZ Myers
105 comments :
Larry,
Have you seen Dan's new book yet? Any assessment?
I'm seeing similarities between assuming the DNA is junk and assuming 'neutral' for gene differences. They are the scientific assumptions.
And I had always thought it was better to assume function.
Well, I was wrong about that, wasn't I?
But now I've got it.
Thanks to Sandwalk blog of all things.
It looks pretty good to me but I'm not really in a position to judge his take on population genetics. He has an entire chapter devoted to eukaryotic genome evolution (82 pages) and he covers all the relevant material.
There's a discussion about the ENCODE fiasco.
The part I disagree with is his parsing of junk DNA into various categories like rubbish DNA, garbage DNA, and indifferent DNA. I don't think that's helpful.
I'm assuming you didn't listen to the talk. Graur isn't assuming DNA is junk. His talk was mostly about the evidence that most of your genome is junk. Toward the end he mentions that, for any specific small bit of that genome, the default assumption should be that it's junk, but that's another matter.
It seems to me that indifferent DNA, at least, is a useful concept. There does appear to be at least a small amount of bulk-function sequence here and there, though the amount in the human genome would seem to have a very low upper limit. You don't think it's nice to have a term for it?
What about removing the chosen 90% from both gametes, create the union and see what develops? You know actually test your concept.
I'm assuming you didn't listen to the talk. Graur isn't assuming DNA is junk. His talk was mostly about the evidence that most of your genome is junk. Toward the end he mentions that, for any specific small bit of that genome, the default assumption should be that it's junk, but that's another matter.
What evidence are you talking about Harshman?
You are not referring to the evidence that is based on bias, evolutionary assumptions are you?
In other words, Graur assumes FIRST the evolution is true-as he sees it-and then proceeds with his imaginary model for it.
The foundation of his bias is that evolution must be true and therefore most of human genome must be junk.
BTW: What about the pufferfish? The population is not large enough to explain it's genome to be almost free of "junk DNA"?
Some DNA may be necessary to separate functional elements; for example, to make a loop. I call it spacer DNA.
There may be a category of functional DNA that just serves to bulk up the genome. I call that bulk DNA.
I prefer a terminology that identifies a function.
Why not a general term to encompass both spacer and bulk DNA, plus any other non-sequence-related functions there might conceivably be? I don't see the harm.
It has been done with mice. Some Darwinists thought the results were positive but vast majority didn't agree.
Population genetics might tell you that 90% is junk but it can't tell you which 90%.
So, Dan Graur is one of the many Jews, like Coyne, who decided to start a war against God because of the Holocaust... That doesn't make evolution true but it makes them to be ferocious about it. Most evolutionists are. You can find out why if you dig deep.
This is a hard experiment to do. It has been done with part of the genome of mice and the mice without presumed junk DNA were fine. This supports the argument that at least some presumed junk DNA really isn't needed.
Of course, more experimentation along these lines would be, probably will be, useful.
The small degree of non-functional DNA in pufferfish can be explained by chromosomal reorganization over long periods of time (aka- evolution). We've seen it before.
What it categorically does not demonstrate is evidence for a "designer".
What you seem to think of as "evolutionary assumptions," hold far more predictive/explanatory power than your philosophy...else, more scientists would find your arguments persuasive.
I might ask, are there any other organisms, besides the pufferfish, that would support your anti-evolution claims?
Don Idiot-xote,
What's derived from the population genetics analyses is the maximum number of genes given the mutation rate. Given those, the enormous amount of DNA in humans could not all be genes or functionally important. It doesn't tell you how much junk there "should" be.
It should be obvious, but of course it isn't to a drooling imbecile like Joe G and a frothing-at-the-mouth anti-Semite like Don Quixote: Doing a knockout experiment involving, say, 90% of a genome requires that one to have first identified all of the functional parts of the genome. Otherwise, if there is only a single regulatory sequence in the deleted segments that is required for survival, the organism dies and you have demonstrated nothing.
Like I say, obvious and simple. But not for these two goofballs.
What a crude, hateful and ignorant statement.
Most atheists are such, for the same reason you don't worship Dionysus, Thor or Bugs Bunny. We see no evidence.
What, pray tell, has you fightin' windmills? The bible teach you that?
Don Idiot-xote,
Oh, and the pufferfish shows that genomes could be much smaller you poor buffoon.
"What about removing the chosen 90% from both gametes, create the union and see what develops?"
As someone here mentioned, removal of non-coding elements without sustaining any noticeable effect could be a matter of timing. They could have a temporary developmental function:
http://www.sciencemag.org/news/2016/08/junk-dna-tells-mice-and-snakes-how-grow-backbone
Whatever the case, it is obvious this debate rages, not in the interest of science, but because complexity is offensive.
Dr. Moran, why are these hateful trolls not banned and their comments allowed through? I come to this blog for educational purposes and seeing comments like this muddying up the discussion is not only inimical but demeaning. Anything you can do to clean this crap up?
While it is true that some DNA segments are used only briefly in development, I believe that in the knock-out experiment the homozygous knock-outs were homozygous as zygotes -- at the one-cell stages.
At about the same time, a professor had his students sequence their own DNA in a region that had small, apparently useless introns. Some students had the introns, others didn't. Lack of introns was not associated with health problems.
In science all knowledge is tentative, but both sets of observations are best explained if some of our DNA is not needed -- is junk.
There was a synthetic biology construction of a yeast chromosome as well. They just left out obvious junk, and it seems to be doing fine.
"Despite all its alterations, yeast cells containing the synthetic chromosome grew just as well as normal yeast. “What’s amazing about it is that there are over 50,000 base pairs that were either deleted, inserted or changed in that chromosome of 250,000 base pairs, and it works. That’s kind of a remarkable effect,” Boeke says."
(Quoted from this pop-press version from Nature - I'm unsure if it is behind a paywall or not).
http://www.nature.com/news/first-synthetic-yeast-chromosome-revealed-1.14941
Leaving it up does serve the purpose of let you know the intellectual and moral voids we are dealing with in these "debates"...
What is it about Larry's blog that brings out these supreme nutcases? Now we have a fucking christian creationist anti-semite. A few weeks ago another nutter was saying he wished for Larry to literally have his flesh set on fire, and before that, that his limbs were amputated. These people are psychotics.
Whatever the case, it is obvious this debate rages, not in the interest of science, but because complexity is offensive.
Riiight. So that article you cited, was it written by Young Earth Creationists?
For God so hated the Jews that he sent his only Son to Earth as one....
The problem with deciding what is junk and what isn't is that you can't test it under all possible conditions. There have been experiments with yeast in which most of the coding genes were deleted one at a time and only about 30% of them appeared to be essential. Investigators went back and deleted them all two at a time and now that number increased. Some genes were only essential when deleted in combination with other genes. That's protein coding regions. Some genes (all highly conserved and all protein coding) still had no obvious function under the limited number of conditions that one can reasonably test in a lab. So I'm not surprised that one can delete a fair amount of non coding sequence in a laboratory yeast strain and still have it grow fine.
I didn’t know the amount of junk had an effect on the mutational load, and still, after going over his talk, do not understand why.
Does it mean that vertebrates with much less junk DNA have to have more offspring to maintain their population size?
Shouldn’t it be instead that it is the amount of functional sites what has an effect on the mutational load?
txpiper: "Whatever the case, it is obvious this debate rages, not in the interest of science, but because complexity is offensive."
No, it's the projection of magic and agency ONTO complexity that is offensive and unwarranted.
The amount of junk constitutes further places where mutations can happen. But the issue is the opposite of your last sentence, the mutational load limits the amount of functional sites. It's what indicated, long time ago, that most of the human genome could not be functionally important.
widgets101,
Yep. According to some studies on complexity (at least as presented to the general public by Kauffman), we cannot expect evolutionary processes to produce organisms with only necessary genes. Degeneracy is to be expected.
Dan's argument is that since DNA replication has a certain measurable error rate and germ cells arise following a certain number of cell divisions in a one-cell embryo, one can calculate on average how many mutations will be present in every sperm or egg. His argument is that if most of the genome is 'functional' then all mutations must be deleterious and if all mutations were deleterious (ie resulted in death of the offspring before that offspring were able to reproduce) then human beings would have to have massive numbers of offspring in order to get enough to survive to reproduce so as to maintain the population. So most of those mutations must not be deleterious- or at least not deleterious to the point of affecting reproductive fitness. Thus, Dan concludes, most of the genome must be 'junk' if you define junk as capable of sustaining a mutation without causing the pre-reproductive death of the organism. My problem with that argument is that because biological organisms arose via evolution and not via optimal design, there's a lot of redundancy in both coding and non coding sequences in genomes. Genome comparisons suggest that vertebrate genomes underwent two complete genome duplications during evolution not to mention many lineage specific duplications of smaller genomic regions. Many genes exist in families such that you have to inactivate more than one family member to produce a phenotype. Does that mean both genes aren't functional? So yes, biological organisms tolerate mutations. They have to for all the reasons Dan points out but it doesn't really address the issue of functionality except from a theoretical POV. The thing I find so silly about the argument Dan has with ENCODE is that they are in essence arguing different things. ENCODE was pretty clear that their definition of function related to biological activity which Dan finds a useless definition because all DNA is replicated and how useful is that. But the biological activities that ENCODE has been annotating are useful because they are biological activities that differ between different cell types at different times in development and between similar cell types in different organisms. They have enabled the identification of elements that regulate where and when individual genes are expressed and it is differences in when and where genes are expressed that is really what evolution, not to mention development and disease to a large extent, are all about. The annotation has also shown that at least some of these biological activities (such as the binding of certain modified histones) are conserved across evolutionary differences over which DNA sequence conservation has been lost.
Dan's analysis doesn't require all mutations to be deleterious. It works if only some proportion are deleterious. The issue is the risk given the proportion of functionally important stuff.
(It's not Dan's analysis, but several people's analyses.)
Paragraphs, widget101. Paragraphs. Perhaps you're unfamiliar with them, but I assure you that using these new-fangled inventions would make it much easier for your readers to figure out what you're trying to say.
ENCODE's original definition of function was NOT biological but chemical. A lot of the DNA did something when tested, but was it of any biological importance? They couldn't know.
"it's the projection of magic and agency ONTO complexity that is offensive and unwarranted."
The frustration obviously has to do with ideological implications. (It's similar to the angry responses towards Schweitzer and Jack Horner when T rex soft tissue was a news item.) Human DNA is supposed to be an evolutionary junkyard. Suggestions that it is not, are not welcome.
The frustration obviously has to do with ideological implications. (It's similar to the angry responses towards Schweitzer and Jack Horner when T rex soft tissue was a news item.)
I don't remember much anger in responses, so much as (understandable) incredulity and skepticism. But the investigators stuck to their guns and produced evidence to support their finding, so it is now accepted.
There are at least two lessons there for you and your fellow creationists.
Human DNA is supposed to be an evolutionary junkyard. Suggestions that it is not, are not welcome.
Yes, which is why ENCODE is being given hundreds of millions of dollars to find function in the genome. Or is all that money coming from creationists?
You're right bwilson - the activities are chemical, not biological and they couldn't know if it was of any biological importance until it was tested. The point is that 1) they defined what they meant by functional and 2) that annotation of the activities they identified has turned out to be of biological importance.
photosynthesis, the argument is that functionality a particular DNA sequence requires that mutation within that sequence will be deleterious and hence if the DNA were 100% functional all mutations would be deleterious.
My apologies for the lack of paragraphs. I rarely make comments long enough to require them but I'll try to do better next time.
True. For that, we have other clues, the best and most easily applied being lack of evolutionary conservation.
Photosynthesis: "The amount of junk constitutes further places where mutations can happen."
I don't see the relationship. We are talking about rates of mutation, doesn't matter how much junk you add functional elements will experience the same number of mutations from one generation to the next.
I understand the rest of what he showed; it's just this bit: There's a slide where he says the larger the amount of junk DNA, the lower the mutation load... May be he just meant to say "the larger we consider the amount of junk DNA in our genome is the lower the mutation load". As stated I was wondering whether two vertebrates with different amounts of junk DNA have different mutational load. It might be my rubbish English...
Lots of exceptions to that John Harshman. Evolutionary conservation doesn't necessarily imply function (at least within every species) and lack of evolutionary conservation doesn't imply lack of function. There are species specific genes and species specific enhancers. Even the detection of evolutionary conservation in non coding regions can be problematic.
Why doesn't conservation imply function? Why can detection of conservation be problematic? I don't say conservation or lack thereof is a foolproof guide to function; there will be exceptions. But I maintain that those exceptions are rare enough to make conservation the best guide we have.
Federico,
I think the assumption there was that genome size is held constant. Perhaps a better way to put it would be that the greater the percentage of junk DNA in a genome of constant size, the lower the mutational load. Clearly, adding more junk DNA to a genome, increasing its size, does zip to mutational load.
Take protein coding regions. Within humans there are a number of genes that have clearly, based on their sequence, arisen through gene duplication. They are referred to as gene families and some of them are quite large. In many cases, if you delete one of those genes (highly evolutionarily conserved), nothing happens. If it's a big enough family of genes (like hox genes) you might have to delete two or three of the genes before you see an effect. So while the genes are functional by any reasonable definition of functionality they can still tolerate inactivation mutations without affecting the reproductive fitness of the organism.
Detection of conservation can be problematic because the proteins that bind to DNA to turn genes on and off generally have very small binding sites (maybe 6-10 bp) and moreover those binding sites are often degenerate so maybe only 4-5 of those 6-10 bp are conserved between individual sites even within a single species. The binding sites themselves often exhibit a lot of redundancy so the regulatory region for a given gene might consist of five or six binding sites spread out over several hundred base pairs of sequence. Maybe only one or two of the sites might be required for activity (and it might not matter which one or two). So what you've got is a region of DNA that is functionally conserved between species (i.e. it acts to regulate the expression of the same gene in different species) but it won't necessarily show sequence conservation in those species.
"Human DNA is supposed to be an evolutionary junkyard."
Creationists only care about humans. And, no, it was never "supposed" to be anything --- this interpretation came after the surprising finding that much DNA in many organisms studed is non-functional.
Thanks, John, I understand it now.
I don't see the relationship. We are talking about rates of mutation, doesn't matter how much junk you add functional elements will experience the same number of mutations from one generation to the next.
If the mutational load is defined as mutations per generation, then more places to mutate (at the same "speed", namely, what you seem to be thinking: mutations per base), means more mutations total per generation.
So you're right, more DNA doesn't change the rate of mutations, thus doesn't change the proportion of mutations in functional places, but the total number of mutations per generation does increase.
Hopefully that came out clear.
No, "annotation of the activities they identified has turned out to be of biological importance" isn't true. Some of it has. Much of it hasn't. Turned out that their chemical definition of "functional" was too broad if one's interest is in DNA sequences of importance to the cells and organisms. They've been walking back their "80% and maybe 100% functional" claim ever since it was made.
Would you agree the annotation itself has been useful in identifying DNA of biological importance?
This sounds like a classic argument between lumpers and splitters in taxonomy. Both can be useful depending on context. Sometimes its better to talk about eukaryotes and not distinguish yeast from euglenoids, sometimes its not.
@widgets101
I'm familiar with all the arguments used by anti-junk scientists to avoid dealing with the evidence against them.
Let's take your gene duplication argument first. I count all genes as part of the functional genome so your argument has nothing to do with the amount of junk in your genome.
Let's take your second argument, which we discussed in class yesterday. What you're saying is that functional regulatory sites will be conserved between species but they might be difficult to detect. They might not even be orthologous in which case the word "conserved" isn't appropriate. ("Converged" is better.")
How much of the genome is functional if you only count short binding sites that are clearly involved in regulation and are present in similar positions in related species? I think you'll find that you are quibbling about a very tiny percentage of the genome.
Sequence conservation is still an important criterion in determining function and lack of sequence conservation (I.e. sites evolving at the neutral rate) is still a good indication that most of our genome is junk.
ENCODE defined what they meant by "functional" then sat by quite happily while others claimed that their results disproved junk DNA. Many of the ENCODE leaders joined in. They are quoted at the time saying that most of the genome is functional, not junk.
A year-an-a-half later they were still waffling when they published their "apology" paper.
Even today, you'd be hard pressed to find an ENCODE leader who is on record saying that most of our genome is junk and most of their biochemically functional sites are spurious noise. Does anyone know of a single example of such an ENCODE leader?
@widgets101
It's been known for over forty years that transcription factors will bind strongly to thousands of sites in the human genome even though only a tiny percentage of these binding sites are actually involved in biologically relevant regulation.
The spurious binding sites will be tissue specific and will be developmentally specific, just like the functional binding sites.
ENCODE has produced massive databases documenting all these spurious binding sites but instead of making a concerted effort to sort out the wheat from the chaff they are concentrating their efforts on expanding the spurious binding site databases.
This does not inspire confidence. Their failure to openly acknowledge that most of the sites have no biological relevance reveals that they don't understand what they're doing. Either that or they're avoiding the truth in order to keep their funding.
Serious question. How do you know that only a tiny percentage of the binding sites for any given transcription factor are involved in biologically relevant regulation and given that there are thousands of sites for hundreds of factors how would you go about sorting out the wheat from the chaff?
There is the option of not taking sides and just saying "we don't know". That is my position.
One issue is that not every nucleotide is absolutely necessary as is. It may have a fine tuning role.
Geneticist George Church gave a presentation at the NIH last week that talked about re-engineered organisms that had as many as 13 codon reassignments! He mentions ENCODE in passing.
What Church shows is that something can live even though a lot of their DNA is totally re-worked, in fact it can even be MORE reproductively successful after having its genome reworked. In other words, function in terms of reproduction can actually be significantly improved by reworking the genome, which then implies that most of a genome may not be functionally optimal in certain dimensions as a matter of principle, which means what defines "function" by ENCODE standards may need to be taken into consideration.
Church's lecture doesn't imply that most of the Genome is junk nor functional, but it does suggest the definition of function needs reconsideration.
There really is a lot we do not know. Church, by the way, gave a quasi endorsement to Stephen Meyer's book.
Here is Church's lecture (with Francis Collins speaking at the start):
https://videocast.nih.gov/summary.asp?Live=21803&bhcp=1
It is worthwhile for any student of biological sciences, and I'm lucky enough to be taking part-time evening courses at the FAES school at the NIH.
Here is a description of the talk:
"The future of genetic codes and BRAIN codes
Dr. Church’s lecture will focus on transformative technologies moving at exponential rates for reading, writing and editing genomes, epigenomes, and other omes. Applications include cells resistant to all viruses via new genetic codes, production and analysis of organs for transplantation, and therapy testing."
https://faes.org/events/wednesday-afternoon-lecture-series-wals-future-genetic-codes-and-brain-codes
There is the option of not taking sides and just saying "we don't know". That is my position.
That position is only reasonable if there are two or more equally viable hypotheses that explain the available observations. Is that the case here? Does the "Most of the genome is functional" model explain the evidence described in Graur's talk? e.g. How does it answer the onion test?
Someone in Graur's talk brought up the subject of structural DNA that is not sequence dependant. I wish Graur has elaborated on this type of function for say, lncRNAs
As an aside, anyone else see the similarity of Graur with Triumph, The Insult Comic Dog? I'll just note that I've never seen the 2 of them in the same place at the same time!
A lot of repetitive DNA is used for structural purposes like insulators for Topologically Associated Domains (TADs). Under varying environmental conditions, it seems some genomes can add a little more protective padding as needed.
Copy number of repetitive elements also has been shown to modulate expression levels.
As Brenda Andrews pointed out, Eukaryotic organisms are highly buffered in terms of their ability to withstand damaging change. Adding a little extra DNA seems one strategy to buffer against damage.
I don't have something immediately handy for onions, but one might extrapolate what is found in mammals to the onion case. Onions can change their expression levels of proteins and RNAs by changing copy numbers in tandem repeats and transposons and hence effect adaptation.
The question is how functional do we classify each nucleotide of such inserted repeats in doing mutational load calculations? Not my problem, it's the problem of people like Graur who obsess over mutational load.
https://academic.oup.com/hmg/article/18/R1/R1/2901109/Copy-number-variants-diseases-and-gene-expression
Uh, right. It'd be a lot simpler, and inspire more confidence in your intellectual honesty, if you just admitted you can't answer the onion test question, without including a bunch of spurious stuff that's just intended to impress me with how well-informed you are.
@widgets101,
We know there are thousands of binding sites. We know that this is exactly what you expect given the shortness of the consensus binding sites. In a large genome there have to be thousands of "random" sequences that match the functional binding site.
We know from studies in bacteria that for every functional binding site there are many spurious binding sites that have no relevant function.
From the data on basic properties of DNA binding proteins and from studies on model cases, we can expect a huge number of spurious binding sites in the human genome and that's exactly what we observe.
Rather than just assume that all these binding sites are functional, as ENCODE does, it would be better to pick one transcription factor and do the necessary biochemistry to show which sites are necessary and which ones are just noise.
You could also do the random sequence experiment proposed by Sean Eddy. That requires inserting a big piece of random sequence DNA into human cells and assaying for transcription factor binding, DNAase I hypersensitivity, histone modification etc. The prediction is that the random DNA will be covered with such markers and much of it will be transcribed.
Larry
How much of the genome do you think is dedicated to building the animal which then becomes inactive when measured after the animal is born? How much of a bacteria's DNA is dedicated to cell division but becomes inactive when measured between divisions?
Would you give me a reference to one of the studies in bacteria to which you're referring because I'm still not sure how you distinguish a spurious site from a relevant site.
I know for example in several mammalian enhancers that have been identified and characterized by mutational analysis that there are often multiple binding sites for a single transcription factor and that you can remove any one or sometimes all but one of them without affecting expression of the gene. So they're all functional but redundant.
Reznikoff, W. S., Winter, R. B., & Hurley, C. K. (1974). The location of the repressor binding sites in the lac operon. Proceedings of the National Academy of Sciences, 71(6), 2314-2318.
Langowski, J., Pingoud, A., Goppelt, M., & Maass, G. (1980). Inhibition of Eco RI action by polynucleotides. A characterization of the non-specific binding of the enzyme to DNA. Nucleic acids research, 8(20), 4727-4736.
von Hippel, P. H., & McGhee, J. D. (1972). DNA-protein interactions. Annual review of biochemistry, 41(1), 231-300.
Maurizot, J. C., & Charlier, M. (1977). Lac repressor binding to single-stranded polyadenylic acid. FEBS letters, 83(1), 107-110.
Allfrey, V. G. (1974). DNA-binding proteins and transcriptional control in procaryotic and eucaryotic systems. Acidic proteins of the nucleus, 2-27.
Yamamoto, K. R., & Alberts, B. (1974). On the specificity of the binding of the estradiol receptor protein to deoxyribonucleic acid. Journal of Biological Chemistry, 249(22), 7076-7086.
Gros, F. (1974). Control of gene expression in prokaryotic systems. FEBS letters, 40, S10-S18.
Chamberlin, M. J. (1974). The selectivity of transcription. Annual review of biochemistry, 43(1), 721-775.
Lin, S. Y., & Riggs, A. D. (1975). The general affinity of lac repressor for E. coli DNA: implications for gene regulation in procaryotes and eucaryotes. Cell, 4(2), 107-111.
Lin, S. Y., & Riggs, A. D. (1972). lac represser binding to non-operator DNA: Detailed studies and a comparison of equilibrium and rate competition methods. Journal of molecular biology, 72(3), 671-690.
Lin, S. Y., & Riggs, A. D. (1970). Lac repressor binding to DNA not containing the lac operator and to synthetic poly dAT.
Lin, S. Y., & Riggs, A. D. (1975). A comparison of lac repressor binding to operator and to nonoperator DNA. Biochemical and biophysical research communications, 62(3), 704-710.
Lin, S. Y., & Riggs, A. D. (1971). Lac repressor binding to operator analogues: comparison of poly [d (AT)], poly [d (A-BrU)], and poly [d (AU)]. Biochemical and biophysical research communications, 45(6), 1542-1547.
Riggs, A. D., Lin, S., & Wells, R. D. (1972). Lac repressor binding to synthetic DNAs of defined nucleotide sequence. Proceedings of the National Academy of Sciences, 69(3), 761-764.
Larry and the "judge-noneses-Dan Graur" are on the mission to challenge God.
Is it fair enough?
If you make the challenge that includes the none-existence of God/ID,
why don't you correct the "errors" and show it how to do it in the first place. I'm sure you can do it otherwise a Designer is going to ...
I would argue that non- specific binding of lac repressor (or any other transcription factor) to synthetic DNA sequence in a test tube isn't relevant because DNA doesn't exist in cells as naked DNA so do any of these studies you cite look at binding in vivo?
Could you try that again, in English?
@widgets101
Do you honestly believe in the face of many decades of studies on DNA binding proteins that every binding site in the human genome is biologically relevant?
You haven't read the papers I gave you. That's not surprising since defenders of ENCODE are remarkably resistant to facts.
If you could only read one paper then I recommend the paper by Keith Yamamoto, a fellow graduate student, and Bruce Alberts, our Ph.D. supervisor. Alberts knows a thing or two about DNA binding proteins and it's time you started learning. You begin by catching up on the literature from 40 years ago.
I don't know if every binding site is biologically relevant. But I don't know that they aren't either because the existence of thousands of different cell types makes it impossible to test. If you operationally restrict biological relevance to those sites whose deletion will cause a phenotype in all genetic backgrounds and in all cell types then, no I don't believe every site will turn out to be biologically relevant. In fact I'm pretty sure that very few would be. Of course only a small proportion of protein coding genes are biologically relevant by that criterion either.
When Salvadoor Cordova (liarsfordarwin) says his position is "I don't know", he's lying. His position is "it's almost all functional" and he's deliberately trying to argue this to his creationist cohorts.
When they come back after having listened to him, believing the genome to be entirely functional, he doesn't correct them and say "you shouldn't believe that, we don't know yet" (although that would also be incorrect), no instead he thanks and praises them for having gotten the point.
To a secular audience he will happily play the Baghdad Bob of creationism, put a giant fake smile on his face and say whatever he thinks he needs to say to feign reasonableness. Yet notice how all his arguments and specious hints are towards the view that it's all functional.
Of course only a small proportion of protein coding genes are biologically relevant by that criterion either.
Really? Citation, please.
Larry
How much of the genome do you think is dedicated to building the animal which then becomes inactive when measured after the animal is born? How much of a bacteria's DNA is dedicated to cell division but becomes inactive when measured between divisions?
Bill, I know you are concerned, but you can rest easy that all DNA required for reproduction is accounted for by selection. In other words, if any form of life can't produce viable offspring, it dies out. Pretty simple, yes?
I am simply humbled that we have Mr. "DNA Steganography" himself here with us. That is, his theory is (or at least was, I can't say whether he still maintains it) that our DNA contains an actual encoded message to the effect that "God built you," something like "Prada" on a designer bag.
That's fine. judmarc. I'm sure Larry Moran, professor of biochemisty, appreciates being reminded by Bill Cole, internet creationist, that living organisms need to reproduce if they are not to die out.
“you can rest easy that all DNA required for reproduction is accounted for by selection…if any form of life can't produce viable offspring, it dies out.”
Perfect. Now, if you could just pin that to the sappy idea of life originating in carbonate rocks riddled with tiny pores, when there was no DNA or anything to be selected, you’d be off to the races.
https://jaysanalysis.files.wordpress.com/2012/05/nano-tech-snake1.jpg
widgets101 says,
I don't know if every binding site is biologically relevant. But I don't know that they aren't either because the existence of thousands of different cell types makes it impossible to test.
Based on what we know about DNA binding proteins, it is unreasonable to think that most transcription factor binding sites are biologically relevant. They certainly aren't in other well-defined systems.
Couple that knowledge with the evidence that 90% of our genome is junk and you'd have to be really stubborn (or a fool) to hold on to the view that most of them are functional.
The megabase pair deletions in mice show that living mice seem perfectly happy without 2Gb of DNA that presumably contain multiple binding sites for many transcription factors. Anti-junk DNA proponents are really clutching at straws when they try to rationalize the results of that experiment.
BTW, how did you like the Yamamoto & Alberts review? Did you follow up on the references and read the papers that looked at transcription factor binding to isolated chromatin vs naked DNA?
Now, if you could just pin that to the sappy idea of life originating in carbonate rocks riddled with tiny pores, when there was no DNA or anything to be selected, you’d be off to the races.
There've been papers written about it, of course, which you're welcome to read and understand.
Me, I'm eagerly awaiting the papers that tell us why several centuries' worth of study by the best ecclesiastical minds in the world wasn't sufficient to understand that God was just being "conversational" when he inspired the writers of the Book of Joshua, and that they should know it's the Earth that moves around the sun. (A footnote explaining how txpiper is much smarter than all those people should be there, too.)
Then there's relativity and quantum mechanics and nuclear physics and the humble physics of the Doppler effect and how to square all of these with a 6000 year old universe - I'm sure you can tell us where all of these ways God's Creation works are mentioned in the Bible, "conversationally" or otherwise, right, tx? But just a paper on the evidence for heliocentrism in the Bible and how all the people studying it prior to Copernicus managed to miss that evidence will be nice for a start.
There is also an inherent issue with ChIP experiments, where highly expressed genes show artifactual enrichment (see Tetylman et al. 2013. PNAS. 110:18602 and Park et al. 2013. PLoS One. 8:e83506). This artifactual enrichment persists even when ChIP'ing nuclear localized GFP (which should have no specific ability to bind DNA). This artifact is not eliminated using untagged controls, which are the standard for ENCODE ChIP experiments. Because different genes are highly expressed in different tissues and under different conditions, differential artifactual binding can masquerade as condition-specific binding. IMHO, these experiments should be used solely for specific hypothesis formulation and testing (e.g. this ChIP peak upstream of my favorite gene is necessary for expression) and not for making broad conclusions (i.e. all of these ChIP peaks must be functional, and since that's not practically testable, let's call it day).
Larry writes:The megabase pair deletions in mice show that living mice seem perfectly happy without 2Gb of DNA that presumably contain multiple binding sites for many transcription factors.
That's very true. On the other hand those same megabase deletions contain a number of very highly conserved DNA sequences which suggests that conservation isn't a reliable of function either.
Then of course being perfectly happy in a laboratory environment isn't necessarily an indicator of how well such mice would do in the wild. Or whether such megabase deletions would still produce functional mice if you made 2Gb worth of deletion in a single mouse.
It's kind of a circular argument. 90% of our genome is 'junk' based on what evidence? Presumably either the mathematical arguments provided by population genetics or the experimental arguments such as the mouse megabase deletions. None of which address the complications in interpretation introduced by genetic redundancy (for which there is also copious experimental evidence).
With all due respect widgets, you are ignoring a large body of model organism research that sheds light on genome functionality. In yeast, 97% of all protein-coding genes show a growth defect when deleted in at least one tested condition (Hillenmeyer et al. 2008. Science. 320(5874): 362-5.). More tractable model organisms (yeast, flies, worms, etc.) are a far better place to start regarding the functionality of presumptive "junk" than mice.
Rumraket:
I don't know for sure how much is functional, but I believe it is mostly functional. Knowledge and belief aren't the same thing.
Larry writes: "The megabase pair deletions in mice show that living mice seem perfectly happy without 2Gb of DNA that presumably contain multiple binding sites for many transcription factors."
That could also be just as good evidence that eukaryotic genomes are "highly buffered" (as in highly robust to injury) to quote Brenda Andrews from University of Toronto at the ENCODE 2015 planning meeting.
For example, each D4Z4 repeat in the Dystrophin gene is about 3.3 kb. Healthy individuals can have 100 tandem repeats of D4Z4. That equates to 330,000 bases of tandem repeats. The region can tolerate at 89% loss of repeats, but if it goes beyond that, it results in serious disease. One could just as well argue the extra D4Z4 tandem repeats FUNCTION to provide deep redundancy. They act as binding sites to all sorts of histone modifying machinery which acts to regulate expression.
50-70% of the genome is composed of repetitive elements. In view of the D4Z4 example, the repetitive elements can also provide the buffering function which Brenda Andrews demonstrated quite powerfully with her high throughput assay experiments on yeast.
A Salty Scientist writes : With all due respect widgets, you are ignoring a large body of model organism research that sheds light on genome functionality. In yeast, 97% of all protein-coding genes show a growth defect when deleted in at least one tested condition (Hillenmeyer et al. 2008. Science. 320(5874): 362-5.).
With all due respect to you Hillenmeyer et al also report that only about 20% of genes exhibit a defect when deleted in cells grown in rich broth (which is more analogous to laboratory mice). Furthermore, laboratory mice although far from perfect are a better model for human beings than yeast are based on the relative proportions of non-coding DNA in the respective species and the number of protein-coding genes.
judmarc,
“There've been papers written about it, of course, which you're welcome to read and understand.“
Nice hop. And your curiosity, of course, ends right there. I’ve read articles about Lane’s speculation, and there isn’t a lot to actually understand because his proposals breach known biological parameters. There is nothing to support magical proto-life formations hanging out in tiny pores till additional miracles happen.
Science functions according to procedures. The empirical approach would be to have repeatable experiments that demonstrate that things like that are possible. At that point, it would be reasonable to look for places and circumstances where they could have occurred without coercion. But that is not how OOL religious foolishness works. A noisy segment of the community just swoons and golf claps at utter nonsense. But they don’t ask for demonstrable evidence, which illustrates that people often believe things on the basis of what they like, or don’t like.
-
“I'm sure you can tell us where all of these ways God's Creation works are mentioned in the Bible, "conversationally" or otherwise, right, tx?”
Of course not. I can’t tell you where the Bible mentions disc brakes or rocket engines, either. If you’re ever curious about atonement, I can help you out with that.
"That could also be just as good evidence that eukaryotic genomes are "highly buffered" (as in highly robust to injury) to quote Brenda Andrews from University of Toronto at the ENCODE 2015 planning meeting."
In short, if the deletion has a phenotype it has a function. If the deletion does not have a phenotype, it has a function. When I grew up that was called 'not doing science'.
Knowledge and belief aren't the same thing.
If you really believed that, you wouldn't be a creationist. (Of the Young Earth variety, moreover, if I understand correctly.)
txpiper,
"And your curiosity, of course, ends right there ..."
That's rich for someone who prefers for curiosities to end at god-did-it.
But, of course not. If that pore thing was satisfactory, we'd agree about how life started and nobody would continue checking. Maybe life started that way, maybe not. The issue is not whether we're satisfied, though, the issue is that you think we should hold to god-did-it. That's not science either.
Science is testing proposals, checking possibilities, etc, until we get some results, then maybe we get more results, maybe we get exited about it, and then we hit a wall, ok then, a proposal that didn't work, let's see elsewhere, etc.
But you don't really give a damn, do you? Even if there was a full fledged answer, you'd continue complaining out of your ignorance, as you surely must do with any other scientific discoveries that conflict with your fantasies.
It would appear that nature has already done this experiment (removing 90% of the genome) for us. The bladderwort genome is just 0.08 billion bases compared to the haploid human genome at 3 billion bases, and the onion genome at 100 billion bases. In just a fraction of the human genome, the bladderwort has 28,000 genes, about the same as humans. We think that about 90% of the human genome is junk, but it appears maybe less than 10% of the bladderwort genome is junk.
So, the answer is in. The bladderwort had most of its junk DNA removed, and it is doing just fine.
his proposals breach known biological parameters. There is nothing to support magical proto-life formations hanging out in tiny pores till additional miracles happen.
If you think anything at all in the peer-reviewed articles on which Nick Lane is an author "breach[es] known biological parameters" or requires anything magical or miraculous, you should alert the relevant journals to a breakdown in their peer review systems. But then, they're all in on the Godless conspiracy to cure cancer and do things in the other sciences like build a better GPS, which I'm sure violates some religious precept of yours.
Science functions according to procedures. The empirical approach would be to have repeatable experiments that demonstrate that things like that are possible
And of course many relevant experiments have been performed and papers published about them, except for the ones that require tens of millions of years; but just like any other science, that which can't be done in the lab is done by careful observation of Nature's "laboratories" if possible. Both the lab experiments and the observations of natural environments show what is presented in these papers is plausible. But scientists, including Nick Lane, will keep trying to figure out things that might be wrong with it, which of course is the biggest difference between science and religion, and the biggest reason science works but religion doesn't as a method of finding out about life, the universe, and everything.
The problem with deciding what is junk and what isn't is that you can't test it under all possible conditions. There have been experiments with yeast in which most of the coding genes were deleted one at a time and only about 30% of them appeared to be essential.
In yeast, 97% of all protein-coding genes show a growth defect when deleted in at least one tested condition.
With all due respect to you Hillenmeyer et al also report that only about 20% of genes exhibit a defect when deleted in cells grown in rich broth (which is more analogous to laboratory mice).
That folks, is how to move the goalposts.
Salty,
I think you have mistaken widgets' thinking. He's considering the yeast example only as a potential support for conclusions drawn from the deletion in lab mice. He's explaining why only the rich broth thing might be relevant. He isn't thinking at all about what the yeast experiment tells us all by itself.
"So, Dan Graur is one of the many Jews...". As Blackadder said: "He's mad! He's mad. He's madder than Mad Jack McMad, the winner of this year's Mr Madman competition."
judmarc,
“But then, they're all in on the Godless conspiracy to cure cancer and do things in the other sciences like build a better GPS, which I'm sure violates some religious precept of yours.”
You can spare me the sanctimony. Do I need to recount how many Christians/creationists have been and are involved in science? Have you ever noticed hospitals with names like Providence, or St. Somebody? Have you ever looked into who very often founded institutions like Johns Hopkins?
The charter for King’s College says that it was founded "for the education of youth in the principles of the Christian Religion, and for their instruction in the various branches of Science and Literature”. It is now the University of Toronto, where Professor Moran works.
-
“And of course many relevant experiments have been performed and papers published about them, except for the ones that require tens of millions of years; but just like any other science, that which can't be done in the lab is done by careful observation of Nature's "laboratories" if possible.”
What nauseating pulp. First, the ‘relevant experiments’ in regards to the origin of life, have barely, if at all, produced anything more impressive than the Miller experiment.
But that aside, it should stand to reason that if something can happen by accident, it should easy to figure out and coerce. If the ideas about precursors were correct, they’d be selling origin of life kits in Walmart.
-
“Both the lab experiments and the observations of natural environments show what is presented in these papers is plausible.”
No, they absolutely do not. You just have an exceptionally low credulity threshold.
"No, they absolutely do not. You just have an exceptionally low credulity threshold"
Says the god-did-it clown.
You can spare me the sanctimony. Do I need to recount how many Christians/creationists have been and are involved in science? Have you ever noticed hospitals with names like Providence, or St. Somebody? Have you ever looked into who very often founded institutions like Johns Hopkins?
And, as usual, the point sails a mile over your pointy little head.
How much of the valuable medical and scientific work done by those institutions have been accomplished by dismissing scientific evidence, and instead taking Holy Scripture as literal truth, even when it is contradicted by reality?
"instead taking Holy Scripture as literal truth, even when it is contradicted by reality? "
As we all know, modern day scientists working with disease models all over the world, use the scientific method described in genesis 30:39 to create new transgenic and knock outs.
They cut out a piece of wood to create knock outs, and add a bit to create transgenics.
Check mate evilutionits.
Heh, 1st comment on the video 'somebodysdad', instantly recognisable as über-berk Joe G, wondering how histone octamers came along subsequently to spool up all the junk ... that being the only way round it could have happened, of course.
Hi Joe, here's a paper about the computational complexity of Ball permutations. I figured your knowledge of complexity in the mathematical realm would be at least equal to your knowledge of complexity in the biological realm, so your comments on this paper would be just as valuable as what you contribute here.
https://arxiv.org/abs/1610.06646
Waiting for your scientific explanation, Allan. Oh wait, you don't have one.
You will never get an answer that satisfies you, as we both know. So I don't know why you keep importuning people across the 'net with these "Explain x to my satisfaction" questions.
Well Joe, time for you to step up to the plate. Do tell how the current diversity of life on this planet did arise? And provide evidence for your claim.
Let me remind you "evolution can't do X, thus goddidit" isn't evidence.
Good luck.
You are just a coward who can only post spamming distractions.
There's a classic projection if I ever saw one.
Joe, here's your answer, in the same "Hillel, explain the entire Torah to me while standing on one foot" spirit in which you asked the question:
Chance and necessity.
Now, to quote Hillel again, "Go and learn it." I mean about biology, because Allan is right: You've displayed your ignorance so repeatedly here that you're accomplishing the exact opposite of impressing anyone.
Perhaps this is the only place where you can hang out and be relatively unique, since you'd qualify as ignorant even among other IDers. Fine, suit yourself, but know we understand how pathetic it is for you to hang out here with the sole purpose of trying to say something stupid enough to get a rise out of people.
with the sole purpose of trying to say something stupid enough to get a rise out of people
Which is why I'm resolving not to respond to Joe G unless he says something relevant (as he did in pointing out that scientists were exploring the idea DNA might have be triple-helical).
I am curious about everyone [in this forum] opinion of this headline and article. Its over my head. http://epigeneticsliteracyproject.org/junk-dna-no-long-non-coding-rna-functional-study-finds/
I cannot judge its work but I take issue with how lncRNA is being used to challenge the junk concept.
Thoughts?
The study suggests there are 20,000 functional lncRNAs. I don't find their data convincing but even if it were true it would only account for 1% of our genome. That still means 90% is junk.
Post a Comment