More Recent Comments

Monday, May 11, 2015

Genomics journal is about to embarrass itself with a special issue on junk DNA

The journal Genomics is a journal devoted to the study of genomes. It describes itself like this ...
Genomics is a forum for describing the development of genome-scale technologies and their application to all areas of biological investigation.

As a journal that has evolved with the field that carries its name, Genomics focuses on the development and application of cutting-edge methods, addressing fundamental questions with potential interest to a wide audience. Our aim is to publish the highest quality research and to provide authors with rapid, fair and accurate review and publication of manuscripts falling within our scope.
They claim that all submissiosn are subjected to rigorous peer review and only 25-30% of submissions are accepted for publication.

The composition of genomes is important so it's no surprise that the journal is interested in publishing articles that address the junk DNA debate. In fact, it is so interested that it is going to devote a special issue to the subject for publication in February 2016.

That's the good news. Now for the bad news ....
Special issue on the functionality of genomic DNAs

Guest Editors:

Prof. Shi Huang
State Key Laboratory of Medical Genetics
Central South University , China
huangshi@sklmg.edu.cn

Prof James Shapiro
Department of Biochemistry and Molecular Biology
University of Chicago
jsha@uchicago.edu

The field of genome evolution and population genetics has for the past half of a century assumed that genomic DNA can be divided into functional and non-functional (“junk”) regions. Experimental molecular science has found little evidence for this assumption. A majority of the noncoding parts of the human genome are transcribed, and numerous experimental researchers have now recognized an important functional role in the so called junk DNA regions, such as syn sites, lncRNA, psudogene transcripts, antisense transcripts, microRNA, and mobile elements. In fact, evidence for functional constraints on noncoding genome regions has long been recognized. New theoretical frameworks based on less arbitrary foundations have also appeared in recent years that can coherently account for the reality of far more functional DNAs, as well as all other major known facts of evolution and population genetics. Nonetheless, there still remains a large gap in opinions between bench scientists in experimental biology and those on the theory side in bioinformatics and population genetics. This special issue will aim to close that gap and provide a view of evidence from a perspective that all genome regions have (or can easily acquire) functionality.

The special issue on the functionality of genome will focus on the following tentative topics:
  1. Theoretical foundation for all genome regions to be functional. It will cover both the theory and all major features of genome evolution.
  2. Functional studies on junk DNA regions, including lncRNA sequences, viral DNAs and mobile elements
  3. Functionalities associated with genome spatial organization in the nucleus
  4. Isocores and compositional constraints on genomes
  5. Genetic basis of complex traits and diseases focusing on the collective effects of normal genetic variations
  6. Cancer genomics
  7. Roles of repetitive DNA elements in major evolutionary transitions
  8. Correlations of genome composition and organismal complexity
  9. Epigenetics
  10. Evo Devo and extended synthesis
Important dates:

First submission date: July 1, 2015
Deadline for paper submissions: October 1, 2015
Deadline for final revised version: December 1, 2015
Expected publication: February 2016
Some of you will recognize the names of the guest editors. Jim Shapiro is one of the poster boys of Intelligent Design Creationism because he attacks evolutionary theory. He's one of the founders of the "The Third Way."

You may be less familiar with Shi Huang. He is also part of the Third Way movement but we've recently learned a lot more about him because he posts comments under the name "gnomon." You can see some of his comments in this thread: Ford Doolittle talks about transposons, junk DNA, ENCODE, and how science should work. Shi Huang appears to have a great deal of difficulty expressing himself in a rational manner.

Those guest editors will publish papers that "... provide a view of evidence from a perspective that all genome regions have (or can easily acquire) functionality." In other words, skeptics need not apply.

The controversy is over the amount of junk DNA in genomes. There are two sides in this controversy. Many scientists think there is abundant and convincing evidence that most of our genome is junk. Other scientists think that most of our genome is functional. It looks like Genomics is only interested in hearing from the second group of scientists. That's why they appointed guest editors with an obvious bias. Those guest editors also happen to be skating very close to the edge of kookdom.

This is not how a credible science journal is supposed to behave.


82 comments :

Anonymous said...

When the ENCODE paper first came out you said you'd be doing damage control for years. Looks like you were right.

T Ryan Gregory said...

I admit that I'm curious who will end up contributing papers...

John Harshman said...

Those guest editors also happen to be skating very close to the edge of kookdom.

You're being much too charitable. They've fallen off the edge, have passed the elephants, and are about to collide with the turtle.

Petrushka said...

There's always a possibility that no sane biologist will submit a paper, considering the editors.

Joe Felsenstein said...

Re: special issue

Hoo boy ...

Petrushka said...

There is always "that" meaning of Special.

I'm sure it will be educational.

Petrushka said...

Missed an opportunity to refer to it as a special educational issue.

Tom Mueller said...

@ Petrushka


There is always "that" meaning of Special.
I'm sure it will be educational.


You did it again! You're killing me!
LOL! LOL! LOL! LOL! LOL! LOL!

... not to mention "Special Olympics"

Jonathan Badger said...

I had never heard of this journal before, which is surprising given that I am a genomicist. I looked it up, and its impact factor is 2.9. By comparison, the two major genomics journals (Genome Research and Genome Biology) have impact factors over 10. If nothing else, one can take solace that the papers in the issue are unlikely to be read or cited much.

Jass said...

I fail to see it as a problem for evolutionary theory. The " Third Way Scientists" still believe in evolution and none of them to my knowledge claim to believe in the first way. Where is the problem if any?

Piotr Gąsiorowski said...

Elsevier is notorious for publishing many dubious or outright pseudoscientific journals. The titles speak for themselves: Journal of Homeopathy, EXPLORE: The Journal of Science and Healing, Chaos, Fractals, and Solitons, etc.

John Harshman said...

The problem is that there's more than one variety of crackpottery, and not all of it is creationism. It's a problem for evolutionary theory if it's invaded by nonsense.

Piotr Gąsiorowski said...

Self-correction: the first title is just Homeopathy. Peer-reviewed (by homeopaths), and with a homeopathic IF.

Joe Felsenstein said...

This will most likely increase their Impact Factor.

Apparently a good way to increase it is to publish a big mistake.

Joe Felsenstein said...

It is worth pointing out to new participants in blog arguments that they should not say "No scientist agrees with you!"

Because a small but noticeable fraction of scientists (I'd say 1%) are self-promoting kooks.

So your opponent can likely find one of those to cite.

Tom Mueller said...

Yeah, something along the lines of a certain BYU physics professor named Steven E. Jones who claimed that the 911 WTC was a Vatican/Zionist/Islamist/al Queda/Mossad/CIA inside job accomplished with demolition bombs.

I understand he too was able to publish his peer-reviewed ravings in a not dissimilar "special educational" journal.

Tom Mueller said...

Here we go

I just found a new site for the Third Way dontchya know.

Some of them were written by PhDs

Diogenes said...

John beats me to it. James Shapiro and Shi Huang are not skating close to the edge of kookdom. They're kooks, and this "special edition" has to exclude contrary evidence and pass off kooky speculations as "evidence" because these particular kooks have zero tolerance for criticism of their psychotic ravings.

Diogenes said...

Jibbers Crabst, you're going to publish a special issue on Junk DNA and the two editors you pick are Shi Huang and James Shapiro? It's like publishing a special issue on all the evidence proving Obama is a Muslim with guest editors Ted Cruz and the gun nut that lives at the end of the road.

And Huang and Shapiro get to reject those who present evidence that Junk DNA exists? Yes, science in the past has always progressed by excluding evidence against a hypothesis. Look at all the great success stories where that worked so well before, like Inquisition v. Galileo, or Lysenko v. Darwin. Smashing strategy that.

gnomon said...

Larry and Joe,
Larry:"Many scientists think there is abundant and convincing evidence that most of our genome is junk."

I ask for a second and final time, show us the beef. Just one "convincing evidence" is enough to satisfy my disinterested interest in this junk debate.

Robert Byers said...

If bias is a factor affecting researchers work then this can be a option for complaint by anyone unhappy with results of the work. So creationism can say bias is going and affecting research into origins./ As a option.

Its the conclusions that are determining peoples confidence, and aggressive rejection of confidence, in science research these days and not a common rulebook.
Creationists face this a lot.
If we question evolutionists they say we are questioning scientists and science and very bad.
Then the moment there is a contention amongst them its okay to question and thjat very aggressively.
Hmmm.
We need rules of criticism.

gnomon said...

To be fair, I first show you my single best piece of evidence for the no junk side. It is the genetic equidistance result first found by Margoliash in 1963, which also happens to be the most remarkable result in molecular evolution since it directly triggered the now defunct concept of a universal molecular clock which in turn directly inspired Kimura to invent the Neutral theory (on Junk DNA theory).

Paul McBride said...

Kimura's (1968) primary argument was genetic load, even if he gave a lot of weight to the molecular clock argument after. The molecular clock is not even strictly a requirement of neutral theory - read what Nei has to say about that, for example. Plus King and Jukes (1969) laid out an intense array of biochemical evidence for the neutral theory that stands independently from the popgen arguments.

Jmac said...

Here we go again.... the genetic load fairytale as the best argument for junk DNA

Somebody shoot me!

Tom Mueller said...

@Diogenes

Your oratorical gift for forensic rhetoric amply justifies your Sinopian pseudonym

;-)

Paul McBride said...

Pest--spoken like someone who made it through half a sentence they didn't understand and stopped reading.

Larry Moran said...

Pest says,

Somebody shoot me!

I'm not a violent person but that's very tempting.

Larry Moran said...

gnomon asks,

I ask for a second and final time, show us the beef. Just one "convincing evidence" is enough to satisfy my disinterested interest in this junk debate.

Here: Five things you should know if you want to participate in the junk DNA debate.

Are you a fan of Michael Denton and Intelligent Design Creationists?

Diogenes said...

Thanks Tom.

Diogenes said...

Yes, this molecular equidistance stuff is very Michael Denton. There are a lot of old entries at TalkOrigins about Denton's misunderstandings of basic genetics. It's rather out of date, even among crackpots, so I haven't debated "molecular equidistance" in years.

I have never once seen "molecular equidistance" presented as evidence for junk DNA nor for neutral theory. The neutrality of most mutations is an observational fact, at least in microbes and flies.

Creationists say they don't disagree with observational science, they just disagree with origin science. That was never true; they always wind up trashing what they call observational science.

I don't know if Shi Huang is a creationist, but he has a major problem with observational science.

Still I'd like to ask him two questions.

1. The average human baby has ~130 novel mutations that its parents did not have. Do you think all of those 130 new imitations are NOT neutral? If they're all not neutral, do you think all 130 mutations are what, catastrophic? Intelligent Design creationists like William Dembski, Phillip Johnson etc. say all mutations are catastrophic. Do you think IDers are full of BS on that? Or do you agree with the IDers' "130 new catastrophes per baby born" hypothesis?

2. The African lungfish has 50x more DNA than we do and, while its mutation rate per generation may be lower, it still must have dozens of times more mutations than humans per baby fish. Do you think the African lungfish has dozens of times more catastrophes per baby than humans do? Or dozens of times more adaptive mutations per baby fish? If the latter, they're on their way to becoming a super race.

Question 2 can be repeated for many other species with huge genomes.

If Shi Huang is an opponent of Junk DNA, I expect him not to answer. They never do.

John Harshman said...

I don't think gnomon's problem bears any resemblance to Denton's. His essential contention, as I understand it, is that most pairwise comparisons of protein sequences are saturated, i.e. randomized by multiple hits. And that differences in protein-sequence distances among species arise because more "complex" organisms have reduced numbers of residues free to vary. Thus humans are closer to chimps than orangutans not because a longer time has elapsed since their common ancestor but because chimps and humans are more complex than orangutans and have more invariant residues, thus reducing the maximum possible sequence difference at saturation. Which is crazy, but a different crazy from what you think.

Diogenes said...

Ugh, more stupid Panglossian idealism! So cytochtome C oxidase and dozens of metabolic enzymes have far more constraints in humans and chimps than in... an orangutan?

There goes ALL of molecular biology. Mol. biologists have mutated many proteins to death. We know how tolerant they are to mutation. Huang then has to throw out all the experimental observations ever made in molecular biology. For molecular biologists it's like: woke up, drank coffee, made a mutant, assayed its function. We know how tolerant proteins are to mutation and that's just talking about CODING regions!

Then there's the megabase deletion mouse.

This Panglossian idealism gets more and more bizarre the more you think about it. What about other species that are similar to each other? Is the Japanese tanuki genetically similar to the Chinese tanuki because tanukis are so complex, so they have a lot of evolutionary constraints and HAVE to be similar?

Endless problems can be totted up. Within certain genera of frogs, one species has less DNA than humans and another species has far more, and the difference between the genome sizes within the genus is larger than the human genome. Doesn't Huang's hypothesis mean that huge variations in genomes is proof of ultra-simplicity? If you think frogs are super-ultra-simple, well then what about mammoths?

What Huang is doing, if John's description is accurate, is called "the displacement problem." You say that your hypothesis is the only answer to a question... but in fact your hypothesis just *displaces* the question to another realm where it is inaccessible and cannot be investigated. Why do living things do what they do? "Their 'vital force' causes it." Why does water act the way it does? "Its' 'aquosity' causes it."

Why are humans so similar to chimps? "Functional constraints cause it " Panglossian $%&*ing crackpots.

Piotr Gąsiorowski said...

Panglossian? I believe Molière rather than Voltaire owns the copyright. In Le malade imaginaire an MD candidate explains -- in awful Latin, during his doctoral examination -- that the cause of the sleep-inducing effect of opium lies in its "vertus dormitiva" (dormitive virtue), and that the laxative action of senna and rhubarb is accounted for by their "vertus purgativa". Needless to say, the examination committee are happy with the answers.

John Harshman said...

Diogenes: It's a bit more complicated than that. First, I'm unclear on what he thinks about genome size. He talks almost entirely about protein sequence distances. Second, not all sequences are saturated in all pairwise comparisons. So some low distances actually do result from close relationship. Just not the ones that for various reasons he doesn't like.

Feel free to read some of his papers. It's hard to understand because of the poor writing, but you can struggle through it and occasionally figure out what he's trying to say. As a student of pseudoscience, you might find it interesting.

SPARC said...

Just out of curiosity: Are 18 of the 1014 amino acids formimg the triple helical part of Type I alpha 2 collagen chains which consist of repeats of the sequnece GXY even suitable for phylogenetic comparisons like those alignments gnomon did to disprove the molecular clock hypothesis?

John Harshman said...

Since he doesn't actually do phylogenetic comparisons, they're as suitable as anything else would be. All he does is count up differences, and see which comparisons have more.

Anonymous said...

@John Harshman

Thanks for translating: Gnomonspeak into English.

I really had problems understanding his arguments while skimming his papers. It looked on the surface like pure gibberish. I did not get why he played the "complexity gambit", now it makes sense in making no sense.

I think, when I have nothing better to do I will indeed use his papers to test my bullshit-detection-capabilities.

P.S.
Funny that somewhere in an earlier thread Peter Borger claimed to understand Gnomons "argumentation" and rated it brilliant.

Steve said...

I guess since Larry and Joe haven't done anything original in lightyears, Shapiro and Huang are good to go.

Nothing pisses skeptics off more than opponents that actually make arguments. You can tell by the oscillating dismissive/accusative nature of their responses.

I love the crackpottery rhetoric. You usually find cracked pots because the plants outgrow the pots usefulness.

so if I were Shapiro and Huang, I'd take a bow. With those two at least we'll get the bus movin' again.

With folks Larry, Joe, Diogenes, Harshman, Piotr, etc etc its just same-o sam-o.

........Loud sound! VVVVVVVVVVVrrrrrrrrrrrrrrrrruuuuuuuuuuuuuuuuuuuuMeMaMeMa.

"What's that??!! Oh, thats just Larry revvin' up his motor. "

"You mean the one he's been workin' on all these years?"

"Yep, that's the one. Sounds great but never seems to be able to actually get it to do anything."

"yeah, I know. But Im like that too. That SOUND is sooooo intoxicating."


John Harshman said...

Just so you know, a light year is a measure of distance, not time.

Anonymous said...

gnomon,

Your questions are irrational. You ask why? Well, for obvious starters, you mistake neutral theory with junk DNA, selectively neutral with random, and random with uniform. Just to name a few obvious problems with your requests. Your problems are much deeper than that, but if you pay close attention, you might start unweaving the mess you're entangled into.

Learn a bit better. Actually work hard at understanding and maybe you'll grow out of your own misunderstandings and misguided efforts.

judmarc said...

Pretty funny watching ID creationists fall all over themselves to get behind papers on how *easy* it is to obtain functionality in the genome. Thus evolution must be successful a high percentage of the time, meaning no outside design is needed...oh, wait.

gnomon said...

Larry:
Thank you for answering my request. A few preludes before I get to the main act. First, I asked only one and you give me five. Why don’t you just pick one so that I could use more words on? Not that I have trouble with any one of the five, but if you have to rely on all five to make your case, you are weak because none of them alone stands a chance. A scientific law should explain all things equally well within its relevant domain.

Second, to use the famous outburst of my favorite tennis legend John McEnroe, “you cannot be serious!!” You call these five things evidence??? We all know evolutionary biology is very soft but you could be accused of making it a non-science!! If a theory doesn’t predict a phenomenon and exclude all other possibilities, it cannot claim credit for the phenomenon or claim it as the theory’s evidence. If your theory let you predict precisely the weather conditions of all of the next 7 days, and can do that repeatedly for any week of the year, then you can rightly claim what is actually observed as evidence of your theory. If your theory predicts that both rain and no rain are possible, then you don’t have a theory. Here, the neutral theory doesn’t predict any of the 5 things. Kimura, Nei, Ayala, Felsenstein, etc, none of them did, at least to my knowledge.

Finally, let’s us be clear that the Neutral theory does predict precisely a few things, some of which when examined for short term popgen or microevo events have been positively verified, e.g., drosophila diversifications in the last few million years in the Hawaii islands.

(BTW, my MGD theory fully absorbs ALL of the proven virtues of the Neutral theory for the linear phase of the diversification in a gene sequence.)

Here are two things predicted by the neutral theory according not to me by leading experts of the theory. 1. the substitution rate predicted by the neutral theory is measured in generations. 2. the theory predicts that the clock will be a Poisson process, with equal mean and variance of mutation rate (1, 2).

So there is some truth to the statement by Kimura and Ohta: “Probably the strongest evidence for the theory is the remarkable uniformity for each protein molecule in the rate of mutant substitutions in the course of evolution.” (3). What they did correctly here is to use the test results of their theory’s prediction as their theory’s evidence. But it is nonetheless a far stretch, because 1. The actually observed substitution rate is measured in years but not in generations as predicted; 2. Experimental data have shown that the variance is typically larger than the mean, rather than equal mean and variance as predicted (1, 2).

gnomon said...

But regardless, a universal molecular clock (approximate or not) is simply a fairytale as verified by bench biology and so in turn is the neutral theory. Here our verdict on the neutral theory (when applied beyond its verified limited domain of relevance) is in strict compliance with standard rules in science: if the prediction of a theory is not met by test results (assuming the test is properly done), the theory is falsified, nothing more and nothing less.

Lastly, I came across the equidistance result on my own when examining my favorite gene RIZ1. To give equidistance credit to Denton is unjust. It is Margolish’s work in 1963. Denton has no clues as what it means except that he is right to point out that it is one of the few most astonishing facts of modern science and is a challenge to modern evolution theory. To use the equidistance result to mock any scientist who tries to seek its explanation is to have no real interest in what the reality is. A few of the ID people know of my work such as Cornelius Hunter as I have left a few comments on his blog. (and Denton too who recently wrote a congratulating note to me on my MGD interpretation of the equidistance, and so he is not your typical ID guys) The fact that the ID camp has yet to make a fuss of the MGD theory makes me wonder that the ID movement may not be motivated by a search for a law to explain the past life history in a contradiction free manner. But unfortunately, some extreme enthusiasts on the other side seem to be motivated by protecting Darwinism based-atheism at all cost. I am sort of in a no man’s land in the middle but hopefully things will change perhaps even very fast.

Now let’s get to your five things just for the fun of it.

1. Genetic load. If there is no negative selection for random errors, yes, those new mutations would destroy a lot of the functional sequences. But our recent papers show a new way for negative selection. Random errors collectively over a MGD threshold can cause diseases and cancer. By inference, may even cause very early spontaneous abortions that people may not even notice. A mutation may be neutral when the total aggregate of all mutations is far below the threshold. This is just like computers eventually crash due to long term use and the accumulation of all kinds of small random hits to its parts.(4,5,6)
2. C-value paradox. No, neutral theory is not the only reasonable explanation (speculation to be more precise). Houses can be small or large and so the number of parts involved can also vary from small to large. If you grant some genome sequences can function as structural elements in addition to information carriers, then the paradox actually makes sense. Also, genome as whole is a building part, and this part can vary in size is just like any parts should have an allowable stdev. Along this line, it fully explains why complex mammals have much less variation in genome size relative to lower species such as fish or plants, according to the first axiom of biology.

gnomon said...

3. Modern evolutionary theory. Well, here we go again, the genetic equidistance result. It falsifies both Darwinism and the Neutral theory, since it is the exact opposite of what is being predicted by these theories.
4. Pseudogenes and broken genes are junk. It is no longer even exciting any more to hear a story that these junks have functions. Sure, not all such junks have been shown to be functional but neither are most proteins. And no one doubts all proteins have functions.
5. Most of the genome is not conserved. To use conservation as an index of function is only measuring one of two kinds of sequences, the essential ones for a species physiology that have little to do with adaption to the outside environments. To maintain the long term integrity of the system, such sequences cannot change. For living fossils to be possible, these sequences should be highly stable. On the other hand, sequences involved in adaption to environments must be fast changing because environmental changes are usually fast. Flu viruses escape neutralizing antibodies every few years, and the fast changing sites in these viruses are absolutely critical for their survival but not essential for their physiology. And the essential sites for physiology of the virus do not change much for say 100 year period.

I am sorry if this reply is too long but, Larry, it may be all your fault as you give me 5 things for me to respond to.
Cheers,
Shi Huang


Ref.

1 Pulquerio M J, Nichols R A. Dates from the molecular clock: how wrong can we be? Trends Ecol Evol, 2007, 22: 180–184
2. Ayala F J. Molecular clock mirages. BioEssays, 1999, 21: 71–75
3. Kimura M, Ohta T. On the rate of molecular evolution. J Mol Evol, 1971, 1: 1–17

4. Zhu, Z., Man, X., Huang, Y., Xia, M., Yuan, D., and Huang, S. (2015) Collective effects of SNPs on transgenerational inheritance in Caenorhabditis elegans and budding yeast. Genomics, http://dx.doi.org/10.1016/j.ygeno.2015.04.002

5. Yuan, D., Zhu, Z., Tan, X., Liang, J., Zeng, C., Zhang, J., Chen, J., Ma, L., Dogan, A., Brockmann, G., Medina, E., Rice, A.D., Moyer, R.W., Man, X., Yi, K., Li, Y., Lu, Q., Huang, Y., and Huang, S. (2014) Scoring the collective effects of SNPs: associations of minor alleles with complex traits in model organisms. Sci China Life Sci. 57:876-888. doi: 10.1007/s11427-014-4704-4


6. Yuan, D., Zhu, Z., Tan, X., Liang, J., Zeng, C., Zhang, J., Chen, J., Ma, L., Dogan, A., Brockmann, G., Goldmann, G., Medina,E., Rice, A.D., Moyer, R.W., Man, X., Yi, K., Li, Y., Lu, Q., Huang, Y., Wang, D., Yu, J., Guo, H., Xia, K., and Huang, S. (2012) Methods for scoring the collective effects of SNPs: Minor alleles of common SNPs quantitatively affect traits/diseases and are under both positive and negative selection. arXiv:1209.2911[q-bio.GN]. ( http://arxiv.org/abs/1209.2911)

gnomon said...

From Larry, five things you need to know:
1. Genetic Load
Every newborn human baby has about 100 mutations not found in either parent. If most of our genome contained functional sequence information, then this would be an intolerable genetic load. Only a small percentage of our genome can contain important sequence information suggesting strongly that most of our genome is junk.
2. C-Value Paradox
A comparison of genomes from closely related species shows that genome size can vary by a factor of ten or more. The only reasonable explanation is that most of the DNA in the larger genomes is junk.
3. Modern Evolutionary Theory
Nothing in biology makes sense except in the light of population genetics. The modern understanding of evolution is perfectly consistent with the presence of large amounts of junk DNA in a genome.
4. Pseudogenes and broken genes are junk
More than half of our genomes consists of pseudogenes, including broken transposons and bits and pieces of transposons. A few may have secondarily acquired a function but, to a first approximation, broken genes are junk.
5. Most of the genome is not conserved
Most of the DNA sequences in large genomes is not conserved. These sequences diverge at a rate consistent with fixation of neutral alleles by random genetic drift. This strongly suggests that it does not have a function although one can't rule out some unknown function that doesn't depend on sequence.

Georgi Marinov said...

@gnomon:

For an evolutionary model to be taken seriously, the population genetics has to work. That's an absolute minimal requirement.

But instead of doing the math and showing it to us, you are arguing with analogies That is never a good sign.

Larry Moran said...

gnomon says,

Second, to use the famous outburst of my favorite tennis legend John McEnroe, “you cannot be serious!!” You call these five things evidence??? We all know evolutionary biology is very soft but you could be accused of making it a non-science!!

Yep, he's a kook.

Unknown said...

Who needs junk? If there is so much junk why don't we just remove it?

Ed said...

Gnomon writes a lot but it all boils down to this bit of text:"some extreme enthusiasts on the other side seem to be motivated by protecting Darwinism based-atheism at all cost."

In Russia these days you're called a neo-fascist if you dare to say you're an atheist.
In Turkey (and most muslim states) religious fanatics threaten and are known to assault teachers who teach evolution. This has gone to such extremes nobody dares to teach evolution any more.
In the US (and many xtian countries) religious fanatics believe you sprout horns and walk on goats hooves if you dare to say you don't believe in literal genesis.

And gnomon expects people to take him serious when he writes "some extreme enthusiasts on the other side seem to be motivated by protecting Darwinism based-atheism at all cost."

Ye right. Gnomon is another one of those people, like peer terborg, who have a pet hypothesis with which they want to overthrow Darwin's theory.
But under scientific scrutiny, their pet hypothesis fails epically.

gnomon said...


Felsenstein, J., (1978) Macroevolution in a model ecosystem. American Naturalist. 177-195.

37 years since publication
Cited 7 times according Google Scholar (self-cite 0)

Zero self-cite most often means that the author himself could not even manage to do any follow up work. But nonetheless, Felsenstein thought that this macro-evolutionary model paper of his is one of his most important work. He went on so far as to say that if any of his work will still be read by someone 50 years from now, it will be this paper.

see the Felsenstein interview here starting at 1:05:50 https://www.youtube.com/watch?v=3wO39cm0a2M

Joe Felsenstein, That is inspiring to me or to any scientist for that matter with a taste for creative theory work. Thank you for sharing that thought.

In comparison, it seems that my macro-evolutionary theory paper is doing not too shabby after all.

Huang, S. (2009) Inverse relationship between genetic diversity and epigenetic complexity. Preprint available, Nature Precedings;

6 years since publication
Cited 13 times according Google Scholar (positive cites by others 5, negative cites by others 0, self-cites 8)

Alex SL said...

I don't really understand half of what is going on here, but...

Wait, you consider self-cites to be more important than other people taking up your ideas, and you consider self-cites outnumbering cites by others to be good? In my area there are quite a few people who constantly go "as [their own name] (year) demonstrated, what I believe is obviously correct", but I never thought it demonstrated more than that they couldn't find many other people to cite who agreed with them.

Also, what is Nature Precedings? The name would imply a journal that existed before Nature...

Unknown said...

Here are two things predicted by the neutral theory according not to me by leading experts of the theory. 1. the substitution rate predicted by the neutral theory is measured in generations. 2. the theory predicts that the clock will be a Poisson process, with equal mean and variance of mutation rate (1, 2).

1) You can measure substitution rates with any unit of time you want. You can use generations, but you can just as well use years.
2) You mess this up somewhat. Subsitiution is a poison process, i.e. the time between substitutions is exponentially distributed with a parameter lambda, and the mean of this exponential distribution is 1/lambda while the SD is 1/lambda. This assumes a constant substitution rate. If the substitution rate is not fixed, but for instance follows a log-normal distribution, then the SD of the distribution of times between substitutions will be larger than the mean.

Your reference (2) mentions BEAST, which does implement this idea. So do the alternatives used these days (dpp-DIV, fdpp-DIV, multidivtime, MCMCtree).

Petrushka said...

I'm pretty sure he intended to say parsecs.

John Harshman said...

You mean as in "It's the ship that made the Kessel run in less than 12 parsecs"?

Petrushka said...

Haven't heard that one furlong, long time.

judmarc said...

Been done (junk removal). And the animals to which it was done got along perfectly well without it.

That won't wash forever, though. If every deleterious mutation occurred in a critical system, over time the population would never be able to support the mutational load, and would go extinct. The fact that millions of species have survived over many generations in spite of mutational load is one good reason we know there *is* junk.

Ed said...

Alex, Nature precedings
In particular this bit is interesting:
"It was a place for researchers to share documents, including presentations, posters, white papers, technical papers, supplementary findings, and non-peer-reviewed manuscripts".

It sounds really cool, to have a paper in Nature *coff coff* prec *coff coff* edings *coff coff*...

Diogenes said...

Steve, what is your evidence that Shi Huang has made a scientific argument? You claim he has made a scientific argument, but you present no evidence he did so. Please *prove* that Shi Huang made a scientific argument. I don't think you really understood what Shi Huang wrote, because no one does. So you're bluffing by pretending you understand the science. I call BS.

Since you claim to understand Shi Huang's argument against neutral theory, or junk DNA, why don't YOU explain it coherently? Shi Huang certainly can't explain anything coherently, rationally or scientifically. So YOU prove he made a coherent point by putting it in your own words.

The only part of Huang's argument I understood is his claim that neutral theory requires a *universal, constant* molecular clock, and there is no *universal, constant* molecular clock. At least I understand that claim enough to explain why it's wrong: neutral theory never required a *universal, constant* molecular clock. It only required neutral mutations to dominate selected mutations, so I dismiss Shi Huang's otherwise incoherent blather.

But prove me wrong. Prove that Shi Huang made a scientific argument by putting it in your own words.

Diogenes said...

Unknown, by what mechanism would "we' remove junk? Do you think mutations that randomly *delete* 10,000 or a million bps are often neutral or beneficial? Do you think we (humans or primates, or who's "we"?) should evolve a mechanism that deletes randomly selected stretches of 10,000 or a million bps at once, because a million generations later, such a mechanism might yield an increase in the efficiency of cellular replication by 0.2%? How would that work in primates? Ya think?

John Harshman said...

I don't think you really understood what Shi Huang wrote, because no one does.

I think that, by great effort, I may have come to understand some of it, at least more than you do. Yes, it's nonsense. There is some that's new and some that's true, but the new parts aren't true and the true parts aren't new. The true part is that the protein clock, such as it is, can't be due to neutral evolution. (Even though most substitutions that occur are neutral, most mutations are not.)

It only required neutral mutations to dominate selected mutations

Not even that. Neutral theory says little about what proportion of mutations are neutral. That would be the genetic load argument. Neutral theory just says that, of those portions of the genome evolving neutrally, fixation rate equals mutation rate.

Diogenes said...

OK. So Shi Huang took five swings and five misses, he considered Larry's 5 bits of evidence for junk DNA and Gnomon could not refute ANY of the five. He did make some factually false assertions, e.g. when we say there's too much variation in genomes for it to be all functional, we're only considering 1 of 2 kinds of sequences, says he. Total BS argument, an attempt to evade by excluding observed variations because they vary.

"Well there's two kinds of evidence, the vast amount that proves me wrong and the tiny bit that doesn't. Your mistake was in considering evidence of the first type." Ya got us, Einstein!

And there's his constant refrain that he's disproven neutral theory.

1. Neutral theory says most mutations are neutral.

2. Something something.

3. Therefore, neutral theory predicts a *universal, constant* molecular clock.

Hmm.I think your problem might be in step 2.

Shi Huang does best at issuing constant ad hominem fallacies, appeals to motive, and appeals to authority. All fallacies, but the comical one is where he tries to denigrate Joe Felsenstein by insinuating Joe has a low citation rate. Joe Felsenstein, one of the 50 top-cited scientists of all time, is dismissed by appealing to the superior authority of one who publishes in Science China Life Sciences.

Here's my favorite Shi Huang appeal to motive: "some extreme enthusiasts on the other side seem to be motivated by protecting Darwinism based-atheism at all cost."

Pure appeal to motive because his evidence is shit. Above, Shi was accusing us of defending neutral theory; here he accuses of defending Darwinism. What an idiot! He doesn't know they're different. He probably thinks the junk DNA hypothesis was based on Darwinism, like Casey $%&*ing Luskin!

Can someone please explain to him how neutral theory and Darwinism are NOT the same? I wish he would pick one ad hominem and stick to it.

Piotr Gąsiorowski said...

Documents on Nature Precedings have not been peer-reviewed and, as such, should not be considered "published" works.

...

Since documents on Nature Precedings are not peer-reviewed, they should not be represented in citations or elsewhere as being peer-reviewed. It is a violation of our terms of service to represent non-peer-reviewed documents on Nature Precedings as peer-reviewed for personal gain.


http://precedings.nature.com/site/help

Larry Moran said...

John Haershman says,

Neutral theory says little about what proportion of mutations are neutral. That would be the genetic load argument.

Actually, the essence of Neutral Theory is that a large percentage of all mutations are neutral. Here's the subtitle of Kimura's 1968 paper:

Calculating the rate of evolution in terms of nucleotide substitutions seems to give a value so high that many of the mutations involved must be neutral ones.

Genetic load arguments and the observation of a molecular clock are what gave rise to Neutral Theory—the idea that many, if not most, mutations are effectively neutral.

The subtitle of the King & Jukes paper (1969) is:

Most evolutionary change is proteins may be due to neutral mutations and genetic drift.

Again, the emphasis is on the proportion of mutations that are neutral. That's what Neutral Theory is all about.

Neutral theory just says that, of those portions of the genome evolving neutrally, fixation rate equals mutation rate.

That's standard population genetics stuff. Neutral Theory really is a proposal that neutral mutations are very common and adaptation is much less important than most scientists thought.

Here's a good reference ...

Ohta, T. (2003) Origin of the neutral and nearly neutral theories of evolution. JOURNAL OF BIOSCIENCES-BANGALORE- 28, 371-377
[PDF]

Diogenes said...

D'oh!

Shi Huang, you got some 'splainin to do! It seems you violated the terms of service of Nature Preceding by falesly presenting your "paper" as "published", when Nature Precedings themselves say it wasn't!

Here again is what Shi Huang wrote:

"In comparison [to Joe Felsenstein, one of the 50 top-cited scientists in the world-- Diog], it seems that my macro-evolutionary theory paper is doing not too shabby after all.

Huang, S. (2009) Inverse relationship between genetic diversity and epigenetic complexity. Preprint available, Nature Precedings;

6 years since publication"


Except your "paper" was never "published"! You sneaky son of a gun. Per Piotr, here, to repeat, is what Nature Precedings says:

Documents on Nature Precedings have not been peer-reviewed... should not be considered "published" works.

Hey, at least Joe's paper was actually published! Bwa ha ha.

Did I mention that Joe Felsenstein is one of the 50 top-cited scientists in the world? And you're not.

Shi Huang, is lying "shabby"?

Unknown said...

Sorry, Larry you're wrong here. There's a difference between the percentage of novel mutations that are neutral and the percentage of substitutions that are neutral. Neutral theory holds that most substitutions are neutral, while you claim that most mutations are neutral and that's not quite right. The majority of novel mutations is detrimental, but the majority of substitutions is neutral. Hence King and Jukes talking about "[m]ost evolutionary change", rather than "most novel mutations".

But yes, the equality of substitution and mutation rates was a result that had been known for quite some time, and when Kimura introduced the diffusion approximation it was one of the classic results he recovered, which shows that it was a viable approximation when Ns~0, which is not true for the 2s approximation (which given decent values for s and N you can also recover from the diffusion model).

I'd also add that the amount of polymorphism expected under neutrality differs from that under alternative hypotheses, because given significant selection the variation disappears faster then under neutrality. That's an important observation in line with neutral theory, which is hard to explain otherwise.

Diogenes said...

Simon, I'm not following your distinction between substitution and mutation.

By mutation do you mean substitution or insertion or deletion?

Or is it that by substitution you mean fixation? I'm confused.

Unknown said...

Substitution rate here is the rate at which novel mutations arise and subsequently get fixed. So you calculate it as µ*N*p, where µ is the rate at which the mutation occurs, N the number of haploid copies of the locus in the population and p the probability of fixation. For a neutral allele p=1/N and therefore you obtain µ.
It doesn't matter whether the mutation in question is an indel or a base change (though µ differs on the type of mutation).

John Harshman said...

The majority of novel mutations is detrimental, but the majority of substitutions is neutral.

Are you referring to amino acid replacement mutations in protein-coding sequences here? If not, I can't make sense of your claim. In any genome that's mostly junk (human genomes, e.g.), most mutations are neutral. That's why it's junk.

Diogenes said...

I have the same problem as John now.

The majority of novel mutations is detrimental, but the majority of substitutions is neutral.

Most novel mutations are neutral (excepting possibly indels) in both protein coding and non-coding regions. In protein-coding regions the fraction of deleterious mutations is higher than in non-coding regions, but in both, neutral mutations are the rule, not the exception.

Unknown said...

Nielsen & Yang (2003) found best fit distributions of S=2Ns using phylogenetic approaches that had a lot of weight in the range between -2.5 and -.5. That would mean that most novel mutations are detrimental. Now, this is precisely what Kimuras model of effectively neutral mutations predicts.

I would not expect mutations in regions of junk to always be neutral. We know that there are responses to TEs in the form of RNAs that stop the TEs from inserting additional copies for instance. This also means that a variant TE that doesn't interact with the RNA would be an issue. In the same way, reactivating ERVs might be bad. Or even just a pseudogene that doesn't get translated being translated again having an energetic cost.
Junk is defined by not having a function. That doesn't mean that there are no mutations in junk regions that could transform them to "actively bad" status.
Now, the N&Y paper only did this for viruses and mtDNA, which have somewhat more streamlined genomes than eukaryotes. But it's worth noting that even that is in line with neutral theory.

Nielsen,R. & Yang,R. 2003 "Estimating the Distribution of Selection Coefficients from Phylogenetic Data with Applications to Mitochondrial and Viral DNA" Mol. Biol. Evol. 20:1231–1239,

John Harshman said...

Simon, you're citing a paper on protein-coding DNA. If that's all you're talking about, I agree, and I don't know where Diogenes is getting his claim from.

Certainly it's possible to have a deleterious mutation in junk DNA. But most of them? That's quite a stretch. "I would not expect mutations in regions of junk to always be neutral" is quite a long way away from "The majority of novel mutations is detrimental". Are you backpedaling?

Larry Moran said...

Sorry, Larry you're wrong here. There's a difference between the percentage of novel mutations that are neutral and the percentage of substitutions that are neutral

I'm pretty sure Simon meant fixation when he said substitution. The rest of us think of substitutions as mutations where one base pair is replaced by another.

Let's just consider the functional part of the genome. In some species that's mostly coding region but in more complex species, like humans, the majority is noncoding (but still functional).

If 8% of the genome is conserved then that's 2% as exons and 6% as something else such as genes for functional RNAs, regulatory sites, origins, parts of introns, centromeres, telomeres etc. Does anyone know what percentage of mutations in that 6% of functional regions is neutral? I suspect it's a significant percentage and maybe more that 50%.

With respect to coding regions, that depends on the gene. About 25% of all mutations in coding regions are synonymous and almost all of them are going to be neutral. (There are exceptions.) So, at least 25% of mutations are neutral in coding regions.

In some genes the number of amino acid substitutions in different species is enormous suggesting that many of the mutations causing amino acid substitutions are neutral. In other protein-coding genes as much as 30% of the residues in the amino acid sequence are conserved in all species. I suspect that even in those genes a significant number of mutations that change codons are neutral.

When you take all of the possible mutations in the functional part of the humans genome it's possible that King, Jukes, and Kimura were right and Simon was wrong. Most mutations are neutral. :-)

Does anyone have solid numbers?

Alex SL said...

Ed,

Ye gods, thanks for pointing that out. I was so sure it was a typo in "proceedings" that I never even thought of looking it up, and it turns out it actually exists. Feeling a bit stupid now.

Also, if it is a repository for things before they are properly published, I see what they are trying to say with that weird name.

Piotr Gąsiorowski said...

It was closed down three years ago:
Announcement

See also
Comment 1 (Retraction Watch)
Comment 2 (Jonathan Eisen)

Unknown said...

Does anyone have solid numbers?

AFAIK no one has tried to do a phylogeny based estimation of selection coefficients for whole genomes. There's a more recent study doing coding regions on mitogenomes (Tamuri et al 2012. ~30% of all mutations and ~10% of non-synonymous mutations neutral, but >90% of all and >70% of non-synonymous substitutions neutral). I'd actually like to do this, but so far the reception by potential collaborators has been lukewarm (I'm pretty sure we do have the raw data to produce solid numbers, it's just a bit out of my comfort zone to do by myself. I'm still a paleo guy with a toe dangling in genomics).

I might well be wrong, because as noted, the estimates that exist are mainly dealing with functional regions.

I would like to point out, that I did not flat out state that you were wrong about that, but that you defined neutral theory incorrectly. Neutral theory is about those mutations that do eventually get fixed. If most mutations are neutral then that is consistent with neutral theory, but it is a stronger claim (i.e. there are scenarios where only a minority of mutations are neutral, yet neutral theory is valid).

I would also like to submit that I'm not the only person to use substitution rate in this way:
http://www.nature.com/nrg/journal/v14/n12/box/nrg3564_BX1.html

John Harshman said...

I haven't seen Larry's meaning of "substitution"; I don't know why it would exist, as it's just a synonym for "mutation". (It's an odd mutation that doesn't change the base.) On the other hand, when I see the word used, it's generally in Simon's meaning. Though sometimes it's used to mean "non-synonymous mutation".

Anyone who claims that most mutations in protein-coding DNA are neutral will have to explain why 1st and (especially) 2nd position substitutions (fixations) are so rare compared to 3rd position ones.

Joe Felsenstein said...

I think we have two different meanings of "substitution" in this discussion.

Simon is correct about most mutations being neutral, but that's when you include all genomic DNA. If you restrict attention to all exons, then John is surely right and most mutations are not neutral.

nmanning said...

"6 years since publication
Cited 13 times according Google Scholar (positive cites by others 5, negative cites by others 0, self-cites 8)"

Self-cites 8; positive cites by others, 5. Wow. Super-genius. David Abel likes to cite himself more than anyone else, too. He also boasts about how many citations he has.

nmanning said...

"Funny that somewhere in an earlier thread Peter Borger claimed to understand Gnomons "argumentation" and rated it brilliant."

Of course. Borger is a proud contributor to Creation Minitries International's fantastic book and DVD "Evolution's Achilles Heels". I'm betting he went on at length about how mutations are so random that they are non-random, and how his "work" on one exon of one zinc finger gene outweighs whole-genome comparisons. He is just that full of himself and his religion.

nmanning said...

That Shapiro has hooked up with Huang all but proves what Shapiro is about.

nmanning said...

Shouldn't the people at the journal GENOMICS understand that this:

"...DNA can be divided into functional and non-functional (“junk”) regions. "

Is total bullshit?

nmanning said...

By the way - Huang has also published in Rivista...