Impractical Naturalism of Dan Graur vs. the NIH

I’ll be making a presentation at AM-NAT 2016, and Dan Graur will be the poster boy of impractical naturalism. Below are some things I collected from his websites, some of which I view as highly anti-science. The aim of my presentation isn’t to settle the question of God or no God or ultimate questions of whether godless naturalism is the best description of reality. The goal is to suggest there are some unspoken naturalistic creeds that often take priority over experiments and observations. In a manner of speaking, there are some interpretations of naturalism that actually go against dispassionate examination of how the natural world actually operates.

More than a decade after the French Academy of Sciences awarded a prize to Loius Pasteur for his work disproving spontaneous generation, Haeckel objected to Pasteur’s claims on philosophical (not scientific) grounds. From the abstract of my AM-NAT 2016,

The success of NIH initiatives and various other projects has drawn a bizarre reaction from [some] Methodological Naturalists in a way somewhat reminiscent of Haeckel who said in 1876: “If we do not accept the hypothesis of spontaneous generation, then at this one point in the history of evolution we must have recourse to the miracle of a supernatural creation.” The demise of the theory of spontaneous generation led to substantial advancement of practical scientific and medical knowledge as symbolized by the Pasteurization process. In the end, the metaphysical implications of the spontaneous generation debate were peripheral to direct experiment and observation necessary to settle scientific questions.

Because of the falsification of spontaneous generation, the germ theory of disease moved forward and so did medical science as a result. The philosophy of Haeckel didn’t prevail and medical science moved forward, even though, by Haeckel’s own admission, the falsification of spontaneous generation raises the specter of a highly improbable and atypical set of circumstances that led to the emergence of life. Years later, these sets of events were deemed so unlikely that they were said to be “almost a miracle” (to quote Francis Crick).

In the present day, the possibility of large amounts of biochemical function in the genome is drawing a reaction reminiscent of Haeckel’s words. This possibility has been most prominently associated with the work of the NIH ENCDOE project and its follow-on projects RoadmapEpigenomics and possibly E4 (Enabling Exploration of the Eukaryotic Epitranscriptome). From publicly available sources, I estimate the sum investment by the NIH could approach almost 800 million dollars, and there could easily be follow on projects.

But Dan Graur protested, “If ENCODE is right, evolution is wrong.”

No one knows the figure of what percentage of the genome is actually functional. Even if the genome is not as functional as some ENCODE researchers suggest, until we actually know which parts are and which parts aren’t functional, as a working hypothesis, much like searching for a missing person, we have to assume any place we look could bring us closer to finding what we seek. So even if ENCODE’s promoters are ultimately wrong, how can it be damaging to science at this stage of the game if we assume high functionality as a possibility? As a matter of investigative procedure, a medical researcher has to at least assume the stretch of DNA he may be studying may have significance until proven otherwise or until we heal every disease on the planet.

Some argue, “well all that ENCDOE money could be spent elsewhere.” Like where? Cancer researchers and researchers of autoimmune disease want to have data on biochemical activity in cells — even if the activity is not immediately indicative of function, it may be a good diagnostic. That data is provided in painstaking detail by ENCODE and RoadmapEpigenomics. Here is a sample of data cancer researchers use from ENCODE:

ENCODE at UCSC

This quote from the above University of California website must get under Graur’s skin 🙂

The Encyclopedia of DNA Elements (ENCODE) Consortium is an international collaboration of research groups funded by the National Human Genome Research Institute (NHGRI). The goal of ENCODE is to build a comprehensive parts list of functional elements in the human genome, including elements that act at the protein and RNA levels, and regulatory elements that control cells and circumstances in which a gene is active.

The parts list is not completely known. ENCODE data provides hints which researchers can pursue and then contribute to defining the parts list. Want to know how researchers use ENCODE data? How about watching some video from this 3 day conference in 2015. The links to the videos are below. You’ll see medical researcher after medical researcher detailing how ENCODE, Roadmap and GWAS data are helping to develop understanding and treatments for various conditions:

https://www.encodeproject.org/tutorials/encode-users-meeting-2015/

One tally of the ENCODE consortium indicated there were 442 researchers spread out in 32 research centers around the world including top names like Harvard, Yale, MIT, Stanford….you get the picture. Yet it didn’t stop Graur from calling them “ignoramuses”.

My point of contention with Graur? He’s being presumptuous, he could be wrong, and he’s advocating an approach that is not helpful to advancing medical science. Not to mention his style is more becoming of a professional troll than a serious researcher.

I provide some quotes in his own words. And disturbing to me personally is that his anti-ENCODE crusade (not at all based on biochemical experiments) didn’t stop him from getting elected as a fellow of the AAAS recently.

And now, as promised, quotes from Gruar himself.

Ewan Birney, the person most responsible for the ENCODE fiasco

he became the scientific equivalent of Saddam Hussein.

http://judgestarling.tumblr.com/page/56
http://judgestarling.tumblr.com/page/38

and

the evolution-free philosophy of ENCODE has not started in 2012. The only difference is that Friedrich Vogel was an honest scientist in a world in which disciplines were rigidly compartmentalized. In comparison, no such excuses exist for ENCODE. My only explanation for their continuing existence is that the wannabe ignoramuses, self-promoting bureaucrats, and ol’ fashion crooks of ENCODE are protected from criticism and penalties for cheating by the person who gives them the money. Thus, they can continue to take as much money from the public as their pockets would hold, and in return they will continue to produce large piles of excrement that are hungrily consumed by gullible journalists who double as Science editors.

http://judgestarling.tumblr.com/page/64

I visit the NIH once or twice a week to learn and report about what is going on there. Many of the researchers of non-coding RNA care deeply about their work in alleviating human suffering. I mingled with a lot of ENOCDE researchers in 2015 at the conference. The medical doctors and medical researchers among them especially care deeply about their work, some working at hospitals for decades. I take offense at Graur calling them “wannabe ignoramuses” and “old fashioned crooks”.

Graur at a talk at the Society for Molecular Biology and Evolution did this:

Toward the end of the presentation, he [Graur] showed a photograph of dollar bills taped together in the shape of a toilet paper roll—his view of what ENCODE had achieved with the $288 million spent on the project so far.

http://www.bio.davidson.edu/genomics/309redesigned/papers_current/ethics_2/2014ENCODE_critic.pdf

toilet paper benjamins

On his website he has a photo of his granddaughter with an extended middle finger and the caption:

My granddaughter, Lilla Keshet Graur, gives ENCODE the finger (2013)

http://nsmn1.uh.edu/dgraur/Encode.html

and of an MIT researcher

When Exactly Did @ManolisKellis Become a Crook? …. By late 2012, however, Manolis Kellis was unquestionably one of the main propagandist for that insidious pseudo-science project called ENCODE.

http://judgestarling.tumblr.com/page/79

and finally

I am sick and tired of all the ENCODEites, their apologists, their hired guns, their propagandists, and all the other badly trained technicians who claim that their favorite piece of DNA does not abide by the rules of evolutionary biology.

… If you cannot come with an evolutionary explanation for the evolutionary functionality of your pet molecule, shut up!

To prove that your sequence is functional in a meaningful evolutionary manner, you have show that the sequence is either maintained by purifying selection or by positive selection (or some other type of selection). You cannot base your claims of functionality on performance of a certain biochemical reaction.

http://judgestarling.tumblr.com/page/115

So how does one prove functionality via evidence of natural selection? Suppose one is dealing with an Orphan Gene, or Taxonimically Restricted Gene (TRG) or feature, an autapomorphy or whatever you call it, etc.? In such cases it’s kind of hard to apply evolutionary conservation to infer selection.

Since when should an evolutionary narrative take priority over lab-derived experimental analyses of biochemical activity? Shouldn’t evolutionary theory be constrained by biochemical analysis rather than the reverse? Dan Graur is a poster child for an impractical brand of naturalism that is more driven by ideological prejudice than empirical methods. So what “if ENCODE is right and evolution is wrong”? Is there a problem with that?

80 thoughts on “Impractical Naturalism of Dan Graur vs. the NIH

  1. Admins,

    SORRY! My preview didn’t show the missing “more” tag. Can someone please add it!

    Profuse apologies!

  2. So what you have shown here is that Dan Graur can be nasty and prone to hyperbole in making attacks on people whose views he doesn’t like. What you haven’t shown is that any of his judgments on the science are wrong, or that he’s being any sort of dogmatic or anti-science crusader.

    Until you actually confront Graur’s arguments, you have nothing worthwhile.

  3. By the way, is this a talk or a poster? If a talk, what session is it in? This doesn’t really sound like something that’s appropriate for a scientific meeting, or at least for Am. Nat. Not a lot of peer review of meeting presentations, eh?

  4. What you haven’t shown is that any of his judgments on the science are wrong, or that he’s being any sort of dogmatic or anti-science crusader.

    You ignored this point:

    Graur wrote:

    To prove that your sequence is functional in a meaningful evolutionary manner, you have show that the sequence is either maintained by purifying selection or by positive selection (or some other type of selection). You cannot base your claims of functionality on performance of a certain biochemical reaction.

    So how does one establish unequivocally selection on an orphan gene or orphan feature by evolutionary theory instead of a biochemical experiment?

  5. AM-NAT. That’s the ID-Creationist conference run by the phony “education” Blyth Institute set up by UD owner Barry Arrington isn’t it?

    Real sciencey there Sal, as per usual.

  6. John,

    AM NAT stands for Alternatives to Methodological Naturalism. JohnnyB advertised it here a few times, but maybe not by the acronym.

    http://www.am-nat.org/site/

    I think Methodological Naturalism is the way to go for questions of operational science, but not ultimate questions like the origin of life, origin of the universe, origin of certain biological features — that is better left to real of not directly provable nor testable, and hence cannot fall under direct experiment nor observation.

  7. stcordova: So how does one establish unequivocally selection on an orphan gene or orphan feature by evolutionary theory instead of a biochemical experiment?

    It depends. If it’s an orphan protein-coding gene, you can still try to estimate sequence-conservation by comparing the DNA sequence of the coding region (rather than the amino acid sequences of the protein), which would most likely correspond to a non-coding region in closely related species. Obviously if the whole coding sequence itself is somehow without any homologous sequences in other species it can’t be done.

    But there really aren’t that many truly de-novo protein-coding genes for which no homologous (non-coding) DNA sequence can be detected in closely related species. In fact I’m not aware of a single case in Homo Sapiens and I suspect there are none.

    The protein might be an Orphan (in the sense that no protein product with a homologous amino acid sequence has been detected in other species), but it’s highly unlikely the chromosomal DNA sequence it is expressed from is totally de novo too. Regardless, the fraction of the genome taken up by such orphan genes is probably below 0.5% and I’m going to predict there are zero cases of Ophan protein-coding genes with no chromosomal DNA homologues in other primates that don’t represent post-divergence retroviral protein-coding genes.

  8. stcordova: So how does one establish unequivocally selection on an orphan gene or orphan feature by evolutionary theory instead of a biochemical experiment?

    You could analyze the knock-out.
    Remember, the “biochemical experiments” Graur is denigrating are “functions” such as “DNAse I hypersensitive site”.
    A good reason to study an organism other than H. sapiens

  9. No. According to ENCODE’s nomenclature, it is not. Read the rest of my post:

    Remember, the “biochemical experiments” Graur is denigrating are “functions” such as “DNAse I hypersensitive site”.

  10. Just to add something to my previous post, there is obviously an issue if a purported orfan is functional and selected for, but so recent that conservation can’t be detected because there’s been divergence up until the point of it’s recent selection. In such a case then yes, I think you’d have to do actual biochemical assays, knock-outs and so on.

    With regards to your thread here Sal, I think the question is how Graur would respond to this scenario. He’s been hyperbolic in his pronouncements I very much agree.

  11. What specifically causes one to side with ENCODE and against Graur?

    Try and avoid the use of the word ‘maybe’ in the answer.

    I understand the competing arguments, and I personally see no reason to concur with ENCODE’s definition of function. But suddenly, Everyone But ENCODE Is Wrong. Why? It smacks of Everyone But Shapiro Is Wrong.

  12. There’s also a tendency (on both sides) to view “ENCODE” as monolithic.
    There’s been some great primary research to come out of this awesome project.
    And some regrettable over-concluding.
    Rather than putting on tin hats and digging trenches (as this OP does), we might all be better served by assessing the various claims separately. Enough with the well-poisoning, already!

  13. What specifically causes one to side with ENCODE and against Graur?

    There are two separate issues really. One is the mission of ENCODE (annotate the genome) and the other is the claim of ENCODE in the 2012 paper (80% functionality). I side with both and I’ll explain why below.

    1. ENCODE/RoadmapEpigenomics mission is to make measurements of biochemical activity. The users of those measurements are mostly medical researchers. The medical researchers are the primary “customers”. I showed one example of an ENCODE genome browsing screen. As you can see, the raw data would be of interest to medical researchers as it can be used to establish baselines of what are markers “healthy” activity versus abberant activity. There is no immediate mention of function or non-function.

    http://www.genome.ucsc.edu/ENCODE/

    When I was at ENCODE 2015, not one presenter that I recall even mentioned the word “junk DNA” nor did they make any issue about the fraction of the genome that is functional or non-functional. The only hint of that was medical researcher after researcher mentioned that a very large fraction of aberrant mutations associated with the disease they were studying was in the non-coding regions.

    This intuitive experience by each of the researchers is consistent with a PNAS study co-authored by the former head of the Human Genome Project and current head of the NIH, Francis Collins, which pointed out 90% of Single Nucleotide Polymorphisms associated with heritable diseases are in non-coding regions. Now, strictly speaking this doesn’t mean the non-coding regions are therefore functional (has causal influence), but one can see why it would be difficult to insist to medical researchers otherwise. The assumption of functionality is a good working hypothesis, and it is hard to run away from, even if ultimately false. We won’t settle the question without more research.

    This would make one “side” with ENCODE for the simple fact ENCODE is providing the assistance that medical researchers have sought. The human genome project had large funding from the NIH, and the natural consequence of actually sequencing the genome is was to find out what the genome actually does.

    One ENCODE researcher at ENCODE 2015 said, “the mission of ENCODE is to annotate the genome.” ENCODE has only started the process. Graur said ENCODE money is waste, the project is run by crooks.

    I side with ENCODE project because we’re not going to annotate the genome (describe the function of the parts) unless we do work like what ENCODE is doing.

    An example of ENCODE work was the search for the numerous locations of the vitamin D receptor binding sites spread across the genome. The genome functions as scaffolding for molecular machines. It’s not just about “pervasive transcription”.

    https://www.youtube.com/watch?v=SYfsUygm0LY

    Below is depiction of enhancer scaffolding which cannot be deduced by mere geneome sequencing, evolutionary theory, but really only by biochemical analysis (like 3C, 4C, 5C, 6C, Hi-C, ChiA-PET, etc). A part of ENCODE 2015 was examination and development of analytical tools for those experiments.

    2. The claim of 80% to 100% functionality based on transcription activity. Admittedly that was the claim of a major ENCODE 2012 paper with all the researchers signing their names to it. Graur took serious exception to that claim and in his defense, no one, including Graur and ENCODE, know what the figure is.

    The “80% functional” claim got him on a tirade which resulted in him calling ENCODE researchers “ignoramuses” and “ole fashioned crooks” and call the ENCODE project a fiasco. Maybe if ENCODE researcher didn’t say that and if creationists weren’t jumping in glee from the quotes from the ENCODE community, Graur wouldn’t have gotten involved since it seems to me, the sort of work ENCODE and Roadmap and possibly E4 is pretty important to the “customers” of the data, namely medical researchers.

    So finally, there are two issues:

    1. the actually mission of ENCODE (to annotate the genome). I side with that!

    2. the claims of ENCODE consortium in the 2012 paper that 80% of the genome is functional. I side with it as a working hypothesis, but it has not been formally demonstrated, not even close. The claim could be wrong, but even if wrong, unless we know what parts are functional and what parts are not, it’s a proper working hypothesis for medical researchers until we actually know one way or the other for every stretch of DNA.

    I do agree with Graur on this point, “If ENCODE is right, evolution is wrong.” To which I respond, “so what?” as far as investigating medically relevant facts.

  14. Sal, you are so profoundly confused. Yet you are also so obviously not an idiot, so I have a very hard time entertaining that what you are doing is truly due to confusion, rather than malice. I mean it. I think you are deliberately trying to engage in misinformation now.

    You don’t need an 80% functional “working hypothesis” to search for and annotate the biochemical activity of all the elements that make up the genome. You don’t need to have a veiw on the percentage of the genome that is functional at all, to do that kind of work. Whether they are functional or not is entirely and completely unnecessary when trying to find out “what it does, if anything”. Something can be nonfunctional and implicated in disease, you don’t have to make any assumptions about what the particular stretch of DNA you are working with does, to TRY TO FIND OUT.

    And it wasn’t advertised as a “working hypothesis”, it was advertised as a conclusion.

    And it was only when they were thoroughly scolded for it they started backtracking on the story. It was a disgusting pr move, the kind of thing you’d expect to be found in a popular press article, not the actual publication itself.

    Simply put, you are working as hard as you can to try to ARTIFICIALLY CONSTRUCT this picture where the percentage of function “working hypothesis” is what has somehow led to discoveries of genome function, and that evolutionary biologists are somehow now standing in the way of this because they disagree with that conclusion.
    What you’re doing is historical revisionism.

    The way you are trying to portray it is not what has actually happened. At no point has any function been discovered because somebody came into the work thinking 80% of the genome was “probably functional”. At no point has a potential function been dismissed because somebody thought 90% of the genome is probably junk.

    Your view of the sequence of events here is terribly misinformed and I honestly suspect you know it and are trying to push this manifestly wrong picture deliberately.

  15. Nice picture by the way. I’m sure it feels like it adds some kind of technical authority and expertise to your witnessing testimony to have that figure in there. Bring a fishing reel schematic next time to really clinch it.

  16. DNA_Jock: There’s also a tendency (on both sides) to view “ENCODE” as monolithic.
    There’s been some great primary research to come out of this awesome project.
    And some regrettable over-concluding.

    ENCODE rolled back much of the over-concluding, so the monoliths are mostly on the ID side.

    And unfortunately, on the pop-sci writing side.

    Larry Moran is angry because it is difficult enough for teachers to undo the damage done by pop-sci writers, but damn near impossible to undo the damage done by refereed literature.

  17. And it wasn’t advertised as a “working hypothesis”, it was advertised as a conclusion.

    But even if the conclusion is ultimately false how does this hurt medical investigation? It think by and large the guys who signed on to the statement believe it. I have a hard time viewing a sincere belief like that as worthy of being called an “ignoramus” or “crook”. I accept it as working hypothesis, maybe the ENCODErs think it is forgone conclusion, and I have a hard time seeing why it would be such a travesty to promote the idea except that it gives fodder to creationists.

    I actually take a more neutral stance on the scientific question, not the least reason is I think personally (not scientifically) that parts of the genome have deteriorated since God threw Adam and Eve out of the garden of Eden, so there could be lots of junk.

    To me, that’s what bothers Dan Graur, there is a cultural change in the way the genome is viewed, and it’s not happening because of creationist propaganda but rather laboratory and medical research.

    That’s probably what’s really galling him.

    And it wasn’t advertised as a “working hypothesis”, it was advertised as a conclusion.

    Ok, so I’ll make sure I’ll change my wording at AM-NAT 2016 so as to not offend you. Thanks for the helpful input.

  18. DNA_Jock: There’s also a tendency (on both sides) to view “ENCODE” as monolithic.
    There’s been some great primary research to come out of this awesome project.
    And some regrettable over-concluding.

    That’s one of the most balanced descriptions I’ve read in the comments, except, why is it “regrettable” if ENCODE over concludes? Does it seriously hurt investigation and future funding of investigation? Isn’t it just as premature for Larry and Dan to insist the genome is junk? Universal Common Ancestry could in principle be right and ENCODE also right — at least that’s what most ENCODErs believe anyway.

    What is really “regrettable” is the ENCODE project gave lots of fodder to the Discovery Institute, IDists and creationists.

    As I said, I think there is the potential for lots of junk because of the mutational load problem that Graur, Kimura, Joe Felsenstein, Nobel Prize winner Herman Muller and last but not least creationist John Sanford and the Mendel’s Accountant team have pointed out. The human genome is deteriorating and natural selection does not have the population resources to arrest it.

    The calculations I provided in this link actually agree with Graur’s published works:

    http://www.uncommondescent.com/genetics/fixation-rate-what-about-breaking-rate/

    And Larry Moran agreed even though he tried to make it sound like he disagreed:
    http://sandwalk.blogspot.com/2014/04/a-creationist-tries-to-understand.html

    Sal Cordova is correct that if the deleterious mutation rate is too high, the species will go extinct. We don’t know the exact minimum number of deleterious mutations that have to happen per generation in order to cause a problem. It’s probably less than two (2). It’s probably not as low as 0.5. It should be no more than 1 or 2 deleterious mutations per generation.

    Where Larry and friends over concludes:

    Genetic load arguments have been around for over forty years [Non-Darwinian Evolution in 1969: The Case for Junk DNA]. Back then, they were used to explain that most of our genome is junk and mutations in that part of the genome have no effect. We now know that those arguments were correct and 90% of our genome is junk.

    It’s formally possible most of the genome is functional even though for this to happen, the functionality had to emerge and be maintained outside of the process of selection. Is that so hard to believe?

  19. stcordova: why is it “regrettable” if ENCODE over concludes? Does it seriously hurt investigation and future funding of investigation? Isn’t it just as premature for Larry and Dan to insist the genome is junk?

    Yes, insofar as people believe ENCODE’s overconclusions, it hurts medical research, because resources are FINITE. And no, the majority of the genome IS non-functional. So, 10% functional might be an underestimate; hell it might be as high as 15%, who knows, maybe even 20% (really stretching it here).
    Your condescension is misplaced.
    Your understanding is lacking: you write (yet again):

    The only hint of that was medical researcher after researcher mentioned that a very large fraction of aberrant mutations associated with the disease they were studying was in the non-coding regions.

    1) Were these “non-coding regions” adjacent to or within genes?
    2)You might want to ponder why coding regions might be under-represented amongst “disease-causing” mutations. Read OMIM and get back to me when you’ve figured that one out.

    This intuitive experience by each of the researchers is consistent with a PNAS study co-authored by the former head of the Human Genome Project and current head of the NIH, Francis Collins, which pointed out 90% of Single Nucleotide Polymorphisms associated with heritable diseases are in non-coding regions.

    Holy crap!
    What do you understand is meant by the phrase “associated with” in this sentence? Is there even the vaguest implication that a SNP “associated with” a disease has anything whatsoever to do with the causation of that disease?

    Now, strictly speaking this doesn’t mean the non-coding regions are therefore functional (has causal influence), but one can see why it would be difficult to insist to medical researchers otherwise.

    Thankfully, medical researchers generally do know what the phrase “associated with” means. Have your classes not yet covered GWAS?

  20. Holy crap!
    What do you understand is meant by the phrase “associated with” in this sentence? Is there even the vaguest implication that a SNP “associated with” a disease has anything whatsoever to do with the causation of that disease?

    Apparently I do, and apparently you didn’t read what I wrote, and if you did, you didn’t comprehend it:

    I said:

    this doesn’t mean the non-coding regions are therefore functional (has causal influence),

    Yet you insinuate:

    Have your classes not yet covered GWAS?

    It was covered a lot at ENCODE 2015. But your criticism is due to your misreading of what I said, not my lack of understanding of GWAS — as I demonstrated by pointing out what I actually said:

    I said:

    this [GWAS studies] doesn’t mean the non-coding regions are therefore functional (has causal influence),

    Care to actually criticize what I said versus criticizing arguments I didn’t make?

  21. stcordova:
    John,

    AM NAT stands for Alternatives to Methodological Naturalism.JohnnyB advertised it here a few times, but maybe not by the acronym.

    http://www.am-nat.org/site/

    I think Methodological Naturalism is the way to go for questions of operational science, but not ultimate questions like the origin of life, origin of the universe, origin of certain biological features — that is better left to real of not directly provable nor testable, and hence cannot fall under direct experiment nor observation.

    Oh, sorry. I thought you were talking about a scientific conference instead of some creationist feel-good meeting. Never mind, then.

  22. Yet again Sal manages to perpetuate the notion that people (before ENCODE) thought that “non-coding” meant “junk”. He just can’t help it.

  23. stcordova:
    I think Methodological Naturalism is the way to go for questions of operational science . . . .

    Like whether or not a global flood occured sometime in the past 10,000 years?

  24. Yes, insofar as people believe ENCODE’s overconclusions, it hurts medical research, because resources are FINITE.

    So where do you suggest money be spent to understand heritable diseases or diseases with a heritable component?

    Would you think money is better spent on genomic research that shows the genome is junk? How would such a research project proceed?

    Seems to me, the ENCODE money is in the right direction, what you don’t like is the fact a lot of the researchers are starting to think the genome has lots of function, not that their work is useless or unfruitful. But who am I or Dan Graur to say where money should be spent, the medical researchers that want this data I think are in the best position to decide, and decide they have — put money on looking for function, don’t put money on trying to demonstrate absence of function.

    I suspect you know ENCODE/Roadmap is doing some stellar science, you just don’t like the fact it’s changing the cultural perception that the Rube Goldberg constructs in biology (like the diagram above) must be exquisitely put together for the Rube Goldberg machines to make life possible.

    I mentioned case of the HOTAIR non-coding lincRNA which originates from Chromosome 12, hops on a molecular bus, goes to a specific locations on Chromosome 2 and then recruits polycomb complexes to switch the memory state of histones — all this to help regulate the development of skin so that skin at the soles of the feet are different from the eyelids.

    HOTAIR was named that because the researcher’s at first thought an RNA that isn’t expressed above the waist, but only below it must be junk — but they studied it anyway, and the journal that published their work was stunned at how non-coding RNA can be recruited.

    For interested onlookers here is the molecular machine that HOTAIR recruits. The “wires” in the diagram are the DNA being used as a scaffold. The “ncRNA binding” in the diagram is generic for the location where ncRNA’s like HOTAIR bind.

    HOTAIR was celebrated example at ENCODE 2015, and here is the molecular complex where HOTAIR parks on chromosome 2 after migrating from chromosome 12 being propelled almost magically on the winds of Brownian motion.

    So what diagrams and experimental research can Dan Graur provide to demonstrate DNA is mostly junk? None that I know of?

    But ENCODE and friends (outside of ENCODE) keep delivering all this data that indicate function, plus lots of pretty pictures of how the genome actually functions — that’s what bothers Graur, not that ENCODE doesn’t provide valuable experimental and observational science.

  25. Yet again Sal manages to perpetuate the notion that people (before ENCODE) thought that “non-coding” meant “junk”. He just can’t help it.

    Where in the OP did I say anything to that effect. You’re channeling DNA_jock and criticizing arguments I don’t make.

    It’s almost like a reflexive thing with some of you guys — start arguing and knocking down arguments I never made.

  26. stcordova,

    There are two separate issues really […].

    Yes I know. I specifically asked why one would side with ENCODE on the definition of ‘function’, not why ENCODE exists or whether it did anything worthwhile.

    You didn’t really answer, or if you did, it was lost. Why would one take positive result in an assay as indicative of biological function?

  27. stcordova,

    Where in the OP did I say anything to that effect.

    Here:

    “Many of the researchers of non-coding RNA care deeply about their work in alleviating human suffering.”. These heroic workers in non-coding RNA are set up in opposition to the Graurs and Morans of the world who (you seem to say) are interested only in protein. It’s just a matter of being careful with your terms.

  28. Allan:

    Why would one take positive result in an assay as indicative of biological function?

    Extrapolating from prior experience with similar situations. Do you have a problem with that? I mean, don’t evolutionary biologists extrapolate all the time without one iota of actual lab demonstration….

    In the above example of ENCODE at the UCSC browser, there is a line that looks for H3K27Ac — the lysine acetylation mark on histone 3. This usually indicates enhancer action. So even if we have not unequivocally established an enhancer is definitely there, from prior experience we suspect there is function from the enhancer regions actually studied experimentally which were associated with the lysine acetylation mark on histone 3.

    Also in the UCSC browser of ENCODE data, we see tables of differential expression. If there is differential expression between cell lines of an RNA, from experience of actual experimental studies, we begin to think if there is differential expression between cells of RNA, this may mean RNA has function.

    Why would you think it is an accident if the cell is going through a lot of trouble to regulate RNA expression?

    Larry argues a lot of transcription is spurious. Well, a lot of transcription would happen if there was not a lot of active machinery to silence and repress it. So the inference is that it is repressed for a physically useful reason. This repression in different cells results in differing RNA expression.

    Why is the conclusion of likely function so absurd to you? That seems like an intuitively good and reasonable extrapolation.

    Ok, so I’ve suggested this is intuitive and an extrapolation. But I’d say Graur and Moran are doing way more unreasonable extrapolations than the ENCODE consortium.

    Allan:

    Why would one take positive result in an assay as indicative of biological function?

    Why would one take a negative result except for evolutionary theory which isn’t really an experimental science compared to biochemistry.

    And to quote the late Austin Hughes:

    In the most recent Heredity Podcast, University of South Carolina evolutionary biologist Austin Hughes explains why he doesn’t believe that positive natural selection is always the primary factor driving the spread of adaptive traits in a population. When asked about the problem with positive Darwinian selection, he says, “The problem is there really isn’t all that much evidence that it actually happens to the extent to which it would be needed to explain all of the adaptive traits of organisms.”

    In Hughes’s view, the evolutionary literature has “clogged up science” with purported examples of evidence of positive natural selection. But he thinks that “most of that literature is worthless”

    http://www.evolutionnews.org/2012/01/austin_hughes_m055121.html

    So why then defer to a theory where most of the literature is worthless.

    I think Hughes is referring to his work that was published in Nature

    http://www.nature.com/hdy/journal/v99/n4/full/6801031a.html

    Recent years have seen an explosion of interest in evidence for positive Darwinian selection at the molecular level. This quest has been hampered by the use of statistical methods that fail adequately to rule out alternative hypotheses, particularly the relaxation of purifying selection and the effects of population bottlenecks, during which the effectiveness of purifying selection is reduced. A further problem has been the assumption that positive selection will generally involve repeated amino-acid changes to a single protein. This model was derived from the case of the vertebrate major histocompatibility complex (MHC), but the MHC proteins are unusual in being involved in protein–protein recognition and in a co-evolutionary process of pathogens. There is no reason to suppose that repeated amino-acid changes to a single protein are involved in selectively advantageous phenotypes in general. Rather adaptive phenotypes are much more likely to result from other causes, including single amino-acid changes; deletion or silencing of genes or changes in the pattern of gene expression.

    So why take the inference from “mostly worthless literature” that has
    “clogged up science” rather than a more positive viewpoint that an assay has shown biochemical activity and thus possible biologically significant function.

    I mean, if one is going to criticize ENCODE for making hasty conclusions, shouldn’t the same standard be applied to Dan and Larry for being too hasty?

  29. Allan Miller,

    Good one, but this is the one I was actually thinking of:

    When I was at ENCODE 2015, not one presenter that I recall even mentioned the word “junk DNA” nor did they make any issue about the fraction of the genome that is functional or non-functional. The only hint of that was medical researcher after researcher mentioned that a very large fraction of aberrant mutations associated with the disease they were studying was in the non-coding regions.

    Not in the OP, but still a fine example of Sal making a claim without even noticing that he’s doing it.

  30. John Harshman: Yet again Sal manages to perpetuate the notion that people (before ENCODE) thought that “non-coding” meant “junk”. He just can’t help it.

    Of course he can help it. He’s being deliberately dishonest because he sees some great value in painting this manifestly false picture of history.

  31. Allan,

    John specifically accused me of this:

    Yet again Sal manages to perpetuate the notion that people (before ENCODE) thought that “non-coding” meant “junk”. He just can’t help it.

    Where did I explicitly say:

    people (before ENCODE) thought that “non-coding” meant “junk”.

    Nowhere! Someone else may have said it. Not me. John’s confusing me with some one else.

  32. stcordova: It’s formally possible most of the genome is functional even though for this to happen, the functionality had to emerge and be maintained outside of the process of selection. Is that so hard to believe?

    That would be an absurdity particularly for a creationist who thinks evolution is basically statistically impossible. What would maintain functionality apart from natural selection? Don’t tell me you’re now conveniently accepting it could happen by pure chance?

    On the one had we are endlessly told that evolution can’t create new information because functional mutations are separated by vast seas of nonfunctionality in sequence space, but now suddenly you’re okay with thinking basically the entire genome and thus every single mutation that separates humans from other great apes is a functional mutation with some kind of phenotypic effect that just hasn’t been selected for?

    You’re trying to have 50 cakes and eat them too with that kind of rationalization.

  33. stcordova: Larry argues a lot of transcription is spurious. Well, a lot of transcription would happen if there was not a lot of active machinery to silence and repress it. So the inference is that it is repressed for a physically useful reason. This repression in different cells results in differing RNA expression.

    You do understand that when Larry argues this, there are actual laboratory demonstrations of the principles backing it up? That the mechanism of transcription is actually quite well understood and that accidental binding and transcription initiation is almost unavoidable?

    There are examples of experiments with artificially constructed, random-sequence (and therefore deliberately made to be junk-DNA) that nevertheless still results in transcription with real transcription factors as-if-it-was-purportedly-functional-according-to-the-interpretation-you-would-lean-towards.

    You can’t just pretend these things aren’t known facts.

  34. So Haeckel was like a creationist/IDist, preferring his preconceptions over the evidence.

    What does this have to do with fights over ENCODE and what-not, other than that Sal tries to force it to be so, and tries to rubbish Grauer using standard ID cant?

    I see nothing having to do with “naturalism” in the disputes, rather disagreements over what the evidence shows. But then, I’ve rarely seen Sal defer to evidence in these matters.

    Glen Davidson

  35. Don’t tell me you’re now conveniently accepting it could happen by pure chance?

    [Btw, I saw you over at Graur’s website. You seem to be a fan of his.]

    I don’t, but I was just offering you a naturalistic solution for your dilemma. You seem to accept chance as cure for the OOL problem, so why not for the maintenance problem? I was just trying to help you out of your dilemma.

    And it may be a dilemma, because even if ENCODE is a few percent right, it still goes beyond Muller’s maintenance limit of 1 mutation per individual per generation given the mutation rate is around 100 per individual per generation.

    If 2 out of the 100 mutantions were in functional regions, this would still blow past the Muller limit. And furthermore, just because there is codon degeneracy doesn’t mean there is no functional change if there is a synonymous mutation. So even if ENCODE is only a few percent right, we could say, “if ENCODE is even a few percent right, evolution is worng.”

    From the so-called Saddam Hussein of the ENCODE empire himself, Ewan Birney:

    http://www.scientificamerican.com/article/hidden-treasures-in-junk-dna/

    BIRNEY: People always knew there was more there than protein-coding genes. It was always clear that there was regulation. What we didn’t know was just quite how extensive this was.

    Just to give you a sense here, about 1.2 percent of the bases are in protein-coding exons. And people speculated that “maybe there’s the same amount again involved in regulation or maybe a little bit more.” But even if we take quite a conservative view from our ENCODE data, we end up with something like 8 to 9 percent of the bases of the genome involved in doing something like regulation.

    Thus, much more of the genome is devoted to regulating genes than to the protein-coding genes themselves?

    And that 9 percent can’t be the whole story. The most aggressive view of the amount we’ve sampled is 50 percent. So certainly it’s going to go above 9 percent, and one could easily argue for something like 20 percent. That’s not an unfeasible number.
    ….
    There has been a lot of debate, inside of ENCODE and outside of the project, about whether or not the results from our experiments describe something that is really going on in nature. And then there was a rather more philosophical question, which is whether it matters. In other words, these things may biochemically occur, but evolution, as it were, or our body doesn’t actually care.

    That debate has been running since 2003. And then work by ourselves, but also work outside of the consortium, has made it much clearer that the evolutionary rules for regulatory elements are different from those for protein-coding elements. Basically the regulatory elements turn over a lot faster. So whereas if you find a particular protein-coding gene in a human, you’re going to find nearly the same gene in a mouse most of the time, and that rule just doesn’t work for regulatory elements.

    One can use the ENCODE data and come up with a number between 9 and 80 percent, which is obviously a very big range. What’s going on there? Just to step back, the DNA inside of our cells is wrapped around various proteins, most of them histones, which generally work to keep everything kind of safe and happy. But there are other types of proteins called transcription factors, and they have specific interactions with DNA. A transcription factor will bind only at 1,000 places, or maybe the biggest bind is at 50,000 specific places across the genome. And so, when we talk about this 9 percent, we’re really talking about these very specific transcription-factor-to-DNA contacts.

    There see, Saddam saw the light! But Graur would not be happy with Birney saying, “evolutionary rules for regulatory elements are different from those for protein-coding elements”. Such comments will get the “ignoraumus”, “Saddam”, “crook” labels coming from Graur real fast. Reminds of Dawkins saying : “stupid, ignorant, insane, wicked.”

    I really like Birney saying this: “the DNA inside of our cells is wrapped around various proteins, most of them histones, which generally work to keep everything kind of safe and happy.

    What did I say about Epignetic machinery (like histones and DNA methylation) being RAM? I can say now, “histones RAM generally works to keep everything kind of safe and happy.” I got it from the man himself, Ewan “Saddam Hussein” Birney.

  36. stcordova: Care to actually criticize what I said versus criticizing arguments I didn’t make?

    LMAO! This is brilliant!
    Sal quotemines himself thus:

    I said:
    this [GWAS studies] doesn’t mean the non-coding regions are therefore functional (has causal influence),

    when what he actually said was

    Now, strictly speaking this doesn’t mean the non-coding regions are therefore functional (has causal influence), but one can see why it would be difficult to insist to medical researchers otherwise.

    Given that you failed to mention GWAS in that post, it is clear to me that you quite deliberately used the phrase “is associated with” in order to imply a causal relationship.

    stcordova: So where do you suggest money be spent to understand heritable diseases or diseases with a heritable component?

    studying the sequences that affect the disease state. Go read OMIM and get back to me.

    … the medical researchers that want this data I think are in the best position to decide, and decide they have — put money on looking for function, don’t put money on trying to demonstrate absence of function.

    And that’s what they have been doing for the past 50 years. Myself, I spent a decade doing this, ending in 1998…
    Sal, you really are coming across as an ignorant twit.

  37. DNA_Jock:

    I used the word GWAS in the OP:

    You’ll see medical researcher after medical researcher detailing how ENCODE, Roadmap and GWAS data are helping to develop understanding and treatments for various conditions:

    Despite the fact I used the word GWAS in the OP you went on and insinuated I was unware of GWAS:

    Have your classes not yet covered GWAS?

    So how do you account for the fact I used the word GWAS in the OP then if I wasn’t aware of the concept of GWAS?

    Grasping at straws there DNA_jock? 🙂

  38. stcordova,

    Extrapolating from prior experience with similar situations. Do you have a problem with that? I mean, don’t evolutionary biologists extrapolate all the time without one iota of actual lab demonstration….

    I don’t have a problem per se. Extrapolation is hardly forbidden. But if you’ve got biochemists like Moran, and many others, saying that they strongly disagree that positive assay=function, and people who actually took part in the ENCODE project themselves saying they disagree with Birney’s assessment (for ’tis largely from him that this figure comes) while your lab experience is … well, what, exactly? – I’m not sure what you are extrapolating from. How many such assays have you performed?

  39. Here is a description of the major ENCODE lab experiments on the genome.

    For all those detractors arguing ENCODE/Roadmap should have their funding cut because the genome is mostly junk, which of the experiments would you recommend defunding related to the list of datatypes ENCODE provides?

    Which experiments would Dan Graur specifically single out as a waste?

    https://www.encodeproject.org/

  40. stcordova,

    Why would you think it is an accident if the cell is going through a lot of trouble to regulate RNA expression?

    It is clear that, in order to determine whether to upregulate something, it has to be examined by the cell machinery. This is simply no guide to function. It is not necessarily accidental, or noise, but part of normal activity. This activity does not mean that the sequence examined has a role.

    Why is the conclusion of likely function so absurd to you? That seems like an intuitively good and reasonable extrapolation.

    As I have seen in other arenas, your intuitions are not particularly sound.

    Me: Why would one take positive result in an assay as indicative of biological function?

    Sal: Why would one take a negative result except for evolutionary theory which isn’t really an experimental science compared to biochemistry.

    I don’t know what you are referring to. Surely a negative result in an assay is not directly indicative of anything?

    There are 3 possible things, broadly:
    1) negative result (uninformative)
    2) positive result (sequence nonfunctional)
    3) positive result (sequence functional).

    1 totals 20%, 2 and 3 between them total 80%. But ENCODE provides no means of ascertaining the latter split.

    I don’t see how one is in any position to say that we are looking, throughout 80% of the genome, at 3, and 2 can be completely discounted.

  41. stcordova,

    For all those detractors arguing ENCODE/Roadmap should have their funding cut because the genome is mostly junk […]

    Who has argued this?

  42. Sal, I’m not sure Graur would single out anything as waste, other than the bad and misleading conclusions.

    If you know what he would eliminate, please say so. Otherwise you are being dishonest.

  43. If you know what he would eliminate, please say so. Otherwise you are being dishonest.

    Well, from Dan himself:

    Toward the end of the presentation, he [Graur] showed a photograph of dollar bills taped together in the shape of a toilet paper roll—his view of what ENCODE had achieved with the $288 million spent on the project so far.

    and

    the evolution-free philosophy of ENCODE has not started in 2012. The only difference is that Friedrich Vogel was an honest scientist in a world in which disciplines were rigidly compartmentalized. In comparison, no such excuses exist for ENCODE. My only explanation for their continuing existence is that the wannabe ignoramuses, self-promoting bureaucrats, and ol’ fashion crooks of ENCODE are protected from criticism and penalties for cheating by the person who gives them the money. Thus, they can continue to take as much money from the public as their pockets would hold, and in return they will continue to produce large piles of excrement that are hungrily consumed by gullible journalists who double as Science editors.

    Petrushka said:

    If you know what he would eliminate, please say so. Otherwise you are being dishonest.

    Ok, so I’ll rephrase it.

    Dan Graur thought ENCODE was a waste and that ENCODers are crooks who produce large piles excrement, but he didn’t exactly specify which of the numerous sequencing and measurement experiments were making large piles of excrement.

    ENCODE did those experiments on 150 cell lines. 5C, ChIA-PET, RNA-seq, CHIP-Seq….. amazing technology. What specifically constitutes “excrement” from the data generated by those experiments? Dan doesn’t specify.

    My perception (not fact) — is Dan isn’t really criticizing the actual work, he hates the fact recognized authorities on the genome and epigenome are daring to say the genome and epigenome has more function than what Dan knows can be evolved and maintained by evolution. As Dan said, “If ENCODE is right, evolution is wrong.” That’s what this is about, it’s not about ENCODE’s data, he doesn’t like researchers openly arguing their opinion of high degrees of functionality. But neither does Dan absolutely know there is so little function, so he can’t categorically declare the ENCODErs as spreading falsehood, at most he can only say they are hasty in conclusion, but then he should say the same of himself!

    If they just kept their suspicions to themselves, they wouldn’t have been lambasted, but because their comments are fodder for creationists, those statements induce Graur’s reaction. That’s the way it looks to me (not saying my perception is fact, but that’s how I see it).

    I think you’re nit picking, but so as to avoid accusations of dishonesty, I’ll try to phrase things differently in deference to you in my AM NAT 2016 presentation. Thanks anyway for your comment.

  44. If we’re talking motivation, I think the fact that the likes of ENCODE, Shapiro and epigenetics are ‘bigged-up’ so extensively by the Creationist community relates to the mistaken belief that they somehow undermine evolution.

Leave a Reply