The writings and life work of Ed Thorp, professor at MIT, influenced many of my notions of ID (though Thorp and Shannon are not ID proponents). I happened upon a forgotten mathematical paper by Ed Thorp in 1961 in the Proceedings of the National Academy of Sciences that launched his stellar career into Wall Street. If the TSZ regulars are tired of talking and arguing ID, then I offer a link to Thorp’s landmark paper. That 1961 PNAS article consists of a mere three pages. It is terse, and almost shocking in its economy of words and straightforward English. The paper can be downloaded from:
A Favorable Strategy for Twenty One, Proceedings National Academy of Sciences.
Thorp was a colleague of Claude Shannon (founder of information theory, and inventor of the notion of “bit”) at MIT. Thorp managed to publish his theory about blackjack through the sponsorship of Shannon. He was able to scientifically prove his theories in the casinos and Wall Street and went on to make hundreds of millions of dollars through his scientific approach to estimating and profiting from expected value. Thorp was the central figure in the real life stories featured in the book
Fortune’s Formula: The Untold Story of the Scientific Betting System that Beat the Casino’s and Wall Street by William Poundstone.
Poundstone’s book doesn’t actually go into detail what formulas actually work in today’s markets because once something works well, it stops working once everyone else figures it out. Thorp’s only published work on how to make money on Wall Street became obsolete and he had to find new avenues of success with secrets he will take with him to the grave….But for those interested, here is Thorp’s only published book on how to make money on Wall Street. As I said it is now obsolete, but it showcases Thorp’s genius and insight. It used to retail for $300 used on Amazon, but then Thorp offered a PDF copy for free:
BEAT THE MARKET
A Scientific Stock Market System.
So if you want a change of pace from the usual arguments over ID, I offer Thorp’s work and you can skip the rest of what is written below since it is my version of ID inspired by Thorp and Pascal.
Though I had followed Thorp’s work on and off for 10 years, I only recently discovered Thorp’s 1961 article while preparing my own draft of a paper that encapsulates my version of Intelligent Design presented at the Blyth Institute’s “Alternatives to Methodological Naturalism Conference” (AM-NAT 2016). I present to the TSZ readers a draft of a paper that I’m submitting as part of the AM-NAT 2016 conference proceedings. AM-NAT 2016 was a conference organized and mentioned by JohnnyB at TSZ and UD. So if you want to argue ID instead of discuss Thorp’s work, that’s ok too.
I got fascinated by the body of math surrounding expected values partly as a result of Thorp’s work. Because of this body of math, I concluded ID theory has been focused too much on information theory and the 2nd law of thermodynamics, and I’ve argued this is a misguided approach. A more clear cut way to frame the probability arguments is to leverage expected values and the law of large numbers and apply similar mathematical approaches, not the approach laid out by Dembski and his almost intractable conception of specification and CSI.
My approach to the question of ID at the personal and practical level has been more along the lines of Pascal’s wagering ideas than trying to make absolutist assertions about reality. Pascal’s wagering ideas were not limited to the theological questions of heaven and hell, but were originally developed to answer theoretical questions about fair values of wagers in gambling games. His solutions using his notion of “expected value” became foundational in probability and statistics, and the notion of expected (or expectation) values has found its way into the realms of physics, chemistry and finance, etc. I’ve framed ID vs. Naturism debate at a personal and practical level more in terms of what is to be gained if ID is right and what might be lost if ID is wrong and how to move forward in science without formal resolution the question of ID.
In my paper, I focused on a practical (not theological) dimension regarding the NIH’s half-billion dollar research investment into the ENCODE and RoadmapEpigenomics projects. Evolutionary biologist Dan Graur has labeled the ENCODE project leaders “crooks” and “ignormuses” and likened the chief architect of ENCODE, Ewan Birney, to Saddam Hussein. Even money aside, there is an issue of bragging rights as to which group of scientists (ENCODE vs. Gruar) should be praised and which group will have egg on their faces after all the dust settles.
To my surprise, the fight over ENCODE spilled over into a fight over what I thought was a rather innocuous article in the New Yorker that promoted the chromatin-centric viewpoint of the epigenome. I did not realize there was a camp (for lack of a better name, I’ll call them the dinosaurs or the transcription-factor proponents or gene-centrists) that was furious at the chromatin-centrists. ENCODE is not only labeled as promoting an “evolution-free gospel” (verbatim words used by rival scientists in a peer-reviewed publication), but they are also not exactly liked by the gene-centrists for their chromatin-centric viewpoint of the epigenome. Creationists and IDists are more sympathetic to the chromatin-centrists, with the qualification that creationists and IDists in general are more favorable to all forms of non-DNA somatic and transgenerational inheritance mechanisms that may reside outside DNA be it organelle, structural, glycomic or whatever “-omic” inheritance devices that may be out there, not just chromatin based mechanisms.
I’ve qualitatively argued the favorable wager in practical terms is on ENCODE, not on evolutionary theory. Most of the paper is rehash of debates I’ve had here at TSZ, so the material is nothing new. It can be said, my paper is really a product of the debates I’ve had at TSZ. The interactions here have helped provide editorial and technical improvements. The paper is still a draft, the figures and formatting will be cleaned up by the myself, reviewers, and the Blyth Institute before it is published in the AM-NAT 2016 conference proceedings. This draft still has a lot of cleanup needed, so I’m posting it to invite improvements. Some of the material might later be reworked as reading material for the high school home school and/or creationist biology students in college. I don’t consider the paper a professional offering, but a way to codify some of my ideas for later reference.
For those tired of reading and arguing what I’ve posted before and have no inclination to read my paper, I provided a link to Thorp’s paper in the chance it may be of passing interest and a change of pace to some readers in this blog.
But for those interested in my paper, here it is:
Gambler’s Epistemology vs. Insistence on Impractical Naturalism: The Unwitting Half-Billion Dollar Wager by the NIH Against Evolutionary Theory.
ACKNOWLEDGEMENTS
Thanks to all at TSZ who have contributed to the refinement of the ideas in my paper. Thanks to the admins and mods of the skepticalzone hosting my postings. TSZ has been a place where I’ve had the chance to receive critical and editorial feedback on materials I’m publishing in various venues.
PS
I had the opportunity to put in practice some of Thorp’s theory in the casino and so did a group of Christians. Their story is documented in the DvD Holy Rollers. I’m listed in the credits of the Holy Rollers documentary.
Here is the trailer. Featured in the trailer is Pastor Mike and other pastors who were part of the Holy Rollers:
I’m not claiming that, at least not now, and if I did in the past, I’ve changed my mind.
I’ve gone on record as saying ID should not be promoted as science.
I accept ID as true however at a personal level.
What ENCODE and RoadmapEpigenomics and the rest of the NIH projects are doing — I consider that science.
stcordova,
I have been researching vitamin d and how low levels increase cancer risk. One hypothesis is that it down regulates the beta catenin protein through transcription of the DKK protein that enables through the WNT pathway the destruction of beta catenin through a 5 protein complex which includes the APC protein. This hypothesis is supported by the APC proteins mutation being implied in many cancers. Beta Catenin is a transcriptional protein for the proteins like MYC that are involved in cell division. Its quantity in the cell nucleus needs to be regulated in order for the tissue to remain healthy(cell division only as required) I am seeing more evidence of other nuclear proteins requiring that their ratios to other proteins be controlled in order for normal cellular function. This ratio also appears to be controlling alternative splicing, which is also implied in many cancers (30000 pubmed papers on this subject). Since the length of introns could control quantities of transcribed proteins I am wondering if their length is relevant to proper cellular function. If introns length is relevant would it still be considered junk because its nucleotide order was not important.
Sal, I just want to point out an annoying thing you do: you ignore most of a post to concentrate on peripheral issues. In this case, you ignore the fact that you have implicitly called for the defunding of evolutionary biology.
colewd,
Hey! Nice to hear from you. Before getting to the bulk of your comment, first off
thank you for your research on vitamin-D.
Since you are the resident specialist on the topic (and this is a tad personal, but I have nothing to lose by talking about it publicly) below was a blood test result that was disconcerting to my doctor regarding my vitamin D levels when I complained of constantly feeling bad. I was at 18 NG/ML when the standard range should be in 30-100 NG/ML. Here was an old test result.
I’m very sensitive to sunlight so I can’t increase my vitamin D by getting sun. Even though I’m dark skinned, I burn easily. Any suggestions?
Thanks for bringing up the topic. Now that I see the mechanical details of how vitamin D affects global gene expression (thousands of genes), it drives home the point how important vitamin D levels are.
I’ve been taking supplements ever since you started the vitamin-D conversation. Perhaps you’ve helped reduce a lot of my future health problems. Any recommended brands and dosages? Thanks.
stcordova,
How long can you spend in the sun before you burn? In 15 minutes you can get a strong dose depending on the season and time of the day. I don’t have any supplement recommendation but talking it is a good idea but orally you can overdose so you need to monitor your blood level. Have your blood level tested again and see if you are moving to the 50ng/ml range.
Repeated question: When are introns considered part of junk DNA?
What the heck does my opinion count in the scheme of things as far as money.
The fine minds of evolutionary biologists and all their comparative techniques should be better deployed in finding patterns of common design rather than supposed common descent, and reframe homology along the lines of Richard Owen who actually coined the term homology to mean a pattern according to a pre-determined plan (common design) rather than common descent.
The mutation load arguments by Graur, Felsenstein, and Muller are accepted by creationists as proof of the infeasibility of universal common ancestry for complex eukaryotic multi-cellular creatures with low reproductive rates. The irony is that Joe (mastermind of evolution) might be also enshrined one day in the creationist hall of fame as well for his work on mutational load, Muller’s ratchet, and mathematical support of genetic ancestry investigations. I highly recommend his book on genetics to my creationist colleagues that are interested in theoretical genetics.
Oddly, some theologically-minded Jews love Joe Felsentstein’s software since it is helping them to argue certain biblical claims. Joe was kind enough to respond to my question whether he was directly involved in the Y-Chromosomal Aaron (Aaron the levitical priest in the time of Moses). For that matter, the Abraham-modal haplotype project, the re-booted mitochondrial eve project, the cattle-eve, the doggie-eve and if they every get around to it tree-eve, turtle-eve etc. would be great evolutionary biology projects for YECs.
If evolutionary biology just redirected its efforts to proving creationism, a world hungry for God would be flooding them with more money than they could ever dream, or at least, they’d grab some share of the hundreds of millions of dollars going to Ken Ham’s amusement parks in Kentucky.
So I’m not necessarily advocating de-funding I’m suggesting re-funding (as in redirected funding) to a different conception of homology.
NOTES:
Joe’s response to my question regarding Y-chromosomal Aaron:
Just show some evidence for that different conception. Like that organisms’ functions are selected by need, not by inheritance (plus change), as we find is the case (imperfectly) in automobiles.
That evidence is lacking, and it matters to science, if not to ID.
Glen Davidson
colewd,
Apologies for deferring your question. I hope you understand I was eager to ask you something relevant to my health. 🙂
In any case:
Both intron length AND sequence could be important, they are not necessarily mutually exclusive.
As mentioned above, histones nucleosome complexes are ubiquitous (1 nucleosome for about every 200 bp of DNA) so even repetitive elements and introns host RAM elements in the form of histones.
Only one experiment, to my knowledge, demonstrated the repetitive nature of some sequences serves as some sort of primitive navigation mechanism for cellular machinery that is not well understood.
Additionally, regarding the Epitranscriptome, when repetitive DNA sequences are transcribed to repetitive RNA sequences, the repetitive RNA sequences could become chemically modified at specific nucleotide positions to actually become quite complex.
If we were to look at the picture of man-made RAM and conclude it wasn’t part of a complex dynamic process merely because the RAM arrays look repetitive, we would be mistaken. In like manner, because of the potential for chemical modifications, the bland looking repetitive RNA sequences coming from bland looking repetitive DNA could function as substrates for RAM modifications in the form of Epitranscriptome chemical modifications. That’s partly why the NIH is throwing around the idea of a 205 million dollar Epistranscriptome research project called E4 (enabling exploration of the eukaryotic epitranscriptome).
In other words, to look at repetitive DNA sequences of variable length and conclude since they look so repetitive, they might be junk would be like looking at the repetitive beads on a string of man-made ram and saying its junk. What makes repetitive God-made RAM and man-made RAM not junk is not their sequential repetitive nature, but how they are modified in real-time dynamically.
Up until the knowledge of the epigenome and epitranscriptome spread, in popular culture DNA (and the associated RNA) were viewed as static entities with fixed read-only memory information — that is clearly not the case, the genome implements ROM and the epigenome RAM and DNA is part of both memory systems.
We really are only beginning to learn how the cell really works, and are chemical techniques are akin to using sledge hammers to break apart and study computer chips.
FWIW,
This is one of the best intron papers I’ve come across. It shows the intron must be looked at from 5 different perspectives, not just DNA sequence, but length, and some other things:
Regarding homology being defined in terms of common design rather than common descent, one can see patterns of similarity without recourse to appealing to common descent, and in fact the assumption of common descent requires a variety of explanations to solve patterns of similarity as shown below.
I’m aware of things like incomplete lineage sorting hypotheses, reductive loss of genes from an ancestor that had more genes than the descendants, etc. But is the assumption of Universal Common Ancestry really needed to see pattern similarities?
In fact, when I spoke of unresolvable phylogenies, what I was referring to is the thought that “sure someone can throw up some algorithm that will give high similarity scores between mice and men, but then gloss over uniquely shared similarities between chicken and men, or mice and chickens.”
http://www.sci-news.com/genetics/article01036.html
No, that isn’t you and the guys at the NIH and ENCODE, it’s just you who uses this absurd reasoning.
Neither is 3 gigabytes. So that’s your entire argument blown out of the water. Even if the entire genome was functional, you STILL wouldn’t have enough storage to specify the location and nature of the TRILLIONS of neuronal connections in your brain. Never mind the trillion cells that make up your entire body.
Which means they aren’t specified in DNA at all, they are the product of an environmentally determined developmental process. The genome contains the information for how to produce a neuronal connection, it doesn’t specificy how or where to make every single one that exists in your brain.
Turns out your “intuitions” about how genetics influence and cause cell-differentiation and communication is completely useless and off the mark.
Onion test. Norwegian sequoia (IIRC) has a genome of 20 gigabases. There are unicellular organisms with 100 gigabase genomes. There are water-lice (smaller than your little finger nail) with genomes ten times larger than humans, and other water-lice species, practically identical, with one-tenth that.
All your rationalizations fail the Onion test.
There is no RAM. It isn’t necessary. You don’t NEED that much storage, you just need a relatively simple program that is environmentally responsive, as in it will run differently if you slightly alter the local circumstances of the cell.
But 10 gigabytes is required for a fucking flea, and 20 for a tree. Gibberish.
The C-value paradox is built on the potentially mistaken assumption most of the information for things like development is in the genome. I’ve provided data points that show heritable developmental information is in the glycome, organelle structure, and who knows what else.
If I gave you some glycan, most likely you can’t show me how DNA codes for it even in principle, ergo, the heritable information for glycans isn’t in the DNA.
DNA specifies sequences for proteins, not information rich glycans.
It’s not an assumption for fucks sake, it’s the product of 50 years of research in genetics and development.
So why do two species of water-lice, practically identical, have an order of magnitude difference in genome size?
Then why the fuck are you hung up on the genome being entirely functional if you are now arguing it is irrelevant to development? Looks to me like you are trying to have your cake and eat it too.
Or maybe I could tell you how enzymes synthesize it, how proteins transport it around and how other enzymes edit it, attach it to proteins, organelles and so on. And then I could tell you how those enzymes are encoded in DNA, how their expression is due to an interaction between the cells environment and the DNA, and so on and so forth. See, I’m not actually claiming everything is “encoded in DNA”, so why the fuck you are bringing up this irrelevant strawman is anybody’s guess.
it also contains transcription data which, in conjunction with environmental cues, determines when, where and how much of all sorts of macromolecules are produced, which in turn affect their mutual interactions and so on and so forth. That doesn’t mean it is true to say it is all “encoded in DNA”, I don’t actually claim that and it is irrelevant to your whole schpiel about ENCODE, evolution, Dan Graur and Junk DNA.
If there’s a lot of structural and developmental information imprinted in intracellular branching carbohydrate chains, then all the more reason why most of the genome can be junk, since it isn’t needed to encode anything then.
But DNA (and its transcripts) aren’t all about encoding proteins. They are part of cellular RAM, not just static unchanging ROM.
The C-value paradox is evidence against the notion that all the developmental information resides solely in the genome, and the fact that glycan structure is not directly templated by DNA but often by pre-existing glycans and other factors is even better evidence lots of developmental information resides outside the DNA, not solely in the DNA.
Glycans are highly important for development. Here are some NIH-funded glycan research projects:
http://authors.library.caltech.edu/65104/
It is interesting to note that in addition to the old tried and true manipulation of glycans via genes, the can be manipulated (mutated) via other means:
and perhaps of some interest to coleWD regarding catenin:
That’s is a pretty good hint of what I just said, developmental information is not constrained solely to the DNA, it’s in the glycans and who knows where else.
Now responding to this:
I didn’t say it is irrelevant to development, it is a necessary but not sufficient component of development.
You’re still not actually dealing with the c-value paradox. Why is there such huge variation even between almost identical species?
Rumraket,
How would evolution by RMNS and genetic drift explain this? How much variation is there?
Sal:
See this:
Things That IDers Don’t Understand, Part 1 — Intelligent Design is not compatible with the evidence for common descent
Take off the God goggles and let the truth wash over you, Sal. It’s refreshing, plus it eliminates raccoon eyes:
As I said in February,
Sorry if I had any role in this latest souce of confusion for poor Sal.
Really! Glycan structure is often templated by pre-existing glycans”.
Do tell!
OTOH, if you meant to say that glycans are made by other factors, such as enzymes, which are “directly templated by DNA”, as Rumraket has pointed out to you, then you would be correct. Your argument would have gone up in smoke too. You need to work on MUCH more specific use of language if you ever want to make an actual, y’know, argument.
Hope springs eternal.
At least ten to one. Read the effing thread.
Mostly transposons, some viral insertions, occasional duplications. That really is it. That’s also one of the reasons we know it’s mostly junk. It’s large swathes of mutationally degraded transposons (most of the junk is). Retrotransposons, transposons that used to be functional (or at least active) but no longer is because mutations have accumulated in them so they can no longer proliferate by inserting themselves elsewhere.
Then there are lots of pseudogenes. These are mostly remnants of viral infections. As in viruses that had parts of their genome inserted in the germline. For example many of these are degraded versions of the virus enzyme Reverse Transcriptase. Having been duplicated over and over again over time.
There are also the occasional pseudogene of a once important functional gene, like the gene that once used to code for the last step in vitamin-C biosynthesis in humans (GULO -> GULOP).
Stuff like that. It’s supported by the genetic load argument, it explains the c-value paradox, and why it also looks like junk when we analyze the sequence. It’s junk, it looks like junk, it “acts” like junk.
Designers can explain any and all patterns, including ones not possible to put into objective nesting hierarchies of shared similarities. You can simply posit that “whatever we see is what the designer wanted to design”. As such it is patently unfalsifiable. Unless you know the mind of the designer beforehand, there is no reason to presuppose a designer will design a nesting hierarchy.
Going into more detail, why do designers some times put things into nesting hierarchies, as in why do they re-use previous designs and modify and derive new ones from them? To save time and resources. Or because of a lack of imagination. None of which applies to the god you believe in. In point of fact, if life really was designed by a materially, temporally and intellectual unconstrained super-intelligence, the breadth of possible life that designer could instantiate would so utterly dwarf what we see it would be absurd to propose our relatively tiny and clearly genealogically related biosphere was the result of such a designer.
ID-creationism fails at every level. It fails scientifically(it makes no predictions, but is only compatible with everything in ad-hoc fashion as “what the designer wanted to design). It fails methodologically(all design-detection methods suggested by ID proponents have been demonstrated to be flawed).
keiths,
Can you explain this in more detail?
Rumraket,
This may all be true but it is not the ID inference according to Behe. This is a straw man argument.
I’m not responding to Behe, I’m responding to the ID inference Sal made and many other ID proponets have made before. It is a response to that and not a strawman of it.
Rumraket,
Thanks for the explanation. Why do you think one specie could collect all this junk and an almost identical specie so much less? Why is there not more genome size variation among other similar species?
Rumraket,
Got it. Thanks
Because it’s random whether deletions, duplications and transpositions happen. A species splits in two, and now both lineages are suffering mutations independently.
In one lineage, by chance the types of mutations that happen include more duplications and some point mutations that deactivate duplicates and transposons.
In the other lineage it suffers a higher frequency of deletions and fewer duplications (say).
Rumraket,
Thanks. Our mitochondrial DNA appears to have very little junk, Do you agree? If you do you have any thoughts why this would have less junk than our genome?
If introns have to be a certain length for the cell to perform properly but the majority of them are just repeating code do you consider the repeating code junk?
I doubt sincerely whether incomplete lineage sorting has anything to do with the pattern in that figure. It would seem to be accounted for wholly by random loss of various genes at various times over a tree of descent. Common ancestry isn’t an assumption, it’s an explanation. It happens to be an explanation that works very well. On the other hand, “that’s just how God designed it” is not an explanation at all. This is why ID (or your creationism) is a science stopper. You can see the pattern, but you can have no explanation for its existence. It’s also a reason why changing biology to a creationist program fails before it begins.
This suggests that you don’t understand the figure you posted, and you don’t understand phylogenetics. “High similarity scores” have nothing to do with how the latter is done. These “uniquely shared similarities” are nothing of the sort; they are artifacts of the taxon sampling: they’re genes that humans and chickens have but mice and zebra fish don’t. If you sampled more taxa you could find exactly the spot on the tree at which that gene arose (somewhere after the zebrafish ancestor diverged from the chicken-mouse-human ancestor, or possibly much earlier and lost in the zebrafish lineage) and where it was lost (somewhere after the mouse ancestor diverged from the human ancestor). That figure is exactly what common descent would predict if there is such a thing as gene gain and gene loss.
To summarize: not only are the phylogenies in question resolvable, that figure doesn’t suggest anything like a lack of resolution. Your ideas of phylogenetics are, if anything, less informed than your ideas of molecular biology, and more dictated by your adamantine but incorrect preconceptions.
Mitochondria are bacteria, and most bacteria have almost no junk. The usual theory is that in tiny, fast-replicating cells, junk has enough of a material and/or time cost that it’s visible to selection.
Do introns have to be a certain length? Not that I can tell. Some genes get by entirely without them. Others have introns that vary wildly in length among closely related species. And the bulk of that length is not usually repetitive sequence.
John Harshman,
It may not be repetitive sequence my point is where intron length may matter but sequence does not. My research in the vitamin d cancer area is revealing that cells are operating with ratios of proteins to remain in equilibrium. Alternative splicing changes when these ratios change. So do functions like cell division. I will email you a PDF of a confirming paper. This makes me wonder if intron length is relevant to cell function because it could moderate transcriptional frequency.
Yes. In fact as far as I can gather mitochondria generally have no junk at all.
This is a bit outside what I know by heart, but afaik the selective pressures operating on mitochondria are much stronger than on the nuclear genome.
They have to be extremely energy-efficient so can’t carry around much(if any) junk, and have most of their volume and surface area devoted to the electron transport complex and ATP production.
They are also a kind of parasite that persists in the genome both because they are advantageous to those that carry them (make lots of ATP), but that also means (since single eukaryotic cells contain multiple mitochondria), in so far as there is more than one of them, if one should mutate they will end up competing and the one that replicates the fastest will dominate. The shorter the genome, the faster it is to replicate it.
Over geological time, most of (what was originally) the mitochondrial genome, has been transferred to the nucleus. What is left is the genes that cannot be transferred because they have to be in close proximity to the electron transport apperatus and ready to be replicated when the host cell splits (and thus needs more mitochondria).
Whole books have been written on this subject. Best one I’ve read is Nick Lane’s The Vital Question: Why is life the way it is?. It really is an amazing book in my opinion. Quite possibly the best one I’ve ever read on anything related to evolution.
That gets into what exactly junk is. You could say that the minimum length, as long as it is necessary for the gene to function, is not junk. But any excess DNA beyond the minimum necessary is just junk repeats.
In that sense, if introns are necessary for a particular gene’s function, then they aren’t junk. To my knowledge, most introns are nonfunctional and could technically be removed with a few modifications here and there. They seem to have proliferated as a kind of selfish DNA that is “parasitic” and have managed to insert themselves in ways that make their presence critical in some genes. As in, in so far as they are there, they need to be removed, and if they are too short for the spliceosome to identify and remove them, they won’t be removed and the host gene won’t function (at least properly).
But the whole thing about them needing to be removed in the first place already argues they are junk. This also implies the spliceosome is an adaptation to their presence.
Well if there is huge variation, then they aren’t exactly identical are they?
That said, I pointed out the importance of the structure of the epigenome in relation to how the genome is used. But first there is an important difference in plant development than animal development:
http://science.sciencemag.org/content/338/6108/758
Now on to your question, the answer from the prestigious journal Science
By golly shazam she listed just about everything I’ve been criticized over regarding my chromatin-centric viewpoint of the epigenome! She even listed the epigenome in terms of histone modifications (ah a closet fan of that hated New Yorker article).
Ah, so duplicated sequences help change gene expression. Nice way to adapt to the environment eh? 🙂
Since the epigenome in plants changes a lot due to environmental factors, it requires changes the genome size to support it.
and
Read it and weep.
keiths, in this OP from 2012:
colewd:
The key is in understanding the difference between the objective nested hierarchy and a merely subjective one.
See this comment for an explanation.
There is no reason to expect an objective nested hierarchy under a hypothesis of common design.
I read it. You didn’t answer my question. Literally everything in that copy-paste speckled with self-gratifying blather is irrelevant to the question posed. I don’t give the slightest fuck about how transposable elements proliferate in animal versus plant genomes, I’m asking you to explain why physiologically, morphologically and ecologically (almost) identical species have such huge variations in genome size. I’m asking you to explain why a microscopic crustacean needs three times more DNA than you do. And why another one that looks exactly like it and lives in a virtually identical environment needs one tenth that much. Repeat that question for literally hundreds of species of plants and animals.
I know an explanation that explains all of it with the same mechanism. In other words, it is a simple explanation for a HUGE collection of data. But now you have to sit there and invent convoluted, case-by-case ad-hoc hypotheses for every such species out there. It’s an absurdity.
Rumraket:
Sal:
That equivocation is pitiful even by your standards, Sal.
Rumraket,
colewd,
The spliceosome: Thats one high tech garbage man. 🙂
Yep. 🙂
You omitted the rest of the sentence changing the intended meaning. I said “other factors”.
I was paraphrasing and extending the implication of a paper I quoted earlier.
The implication is if there is not a direct DNA-template, there must be some other indirect “template” that may involve a host of components.
The host of “environmental” factors also includes the influence on glycans that were influenced by environmental changes which are part of glycoprotein complexes that influence the synthesis of other glycans. I was describing an indirect template rather than one that is more a codon-to-amino acid direct type template as with DNA translation.
You were keen to render the worst possible interpretation of what I say, so I know not to use that choice of words in the future. I thank you on behalf of the creationist kids who’ll someday read my offerings for your free-of-charge editorial review.
Do you still insist chromatin is fuzzy concept despite the fact there are electron micrograph pictures of chromatin in this discussion?
Do you not understand the difference between the fact that the structure of DNA is a necessary condition for glycans to form properly, but DNA is not a sufficient condition for glycans to form properly.
Can one infer in principle from the structure of DNA what the glycan structure should be of all the glycans in a cell that appear transgenerationally? Of course not. That’s because DNA doesn’t contain all the information.
82.5 megabytes of DNA information does not a human make.
Actually I was trying to point out the Rumraket’s oxymoronic statement of “huge variation even between almost identical species”
He probably meant huge gene size variation without much phenotypic effect. I thought the article I quoted provided an excellent answer if one thinks about it.
There is difference in the variants in terms of there ability to cope with possible environmental stress. Sometimes the difference are not noticeable until an environmental stress is present.
Since exposure to environmental stress results in heritable genetic and epigenetic changes and we know that genetic and epigenetic changes don’t necessarily lead to noticeable phenotypic changes it is possible this explains what is going on.
Sal,
Fixed that for you.
colewd,
I’ve looked at the paper, but so far I don’t see what it has to do with intron length. Can you clarify?
John Harshman,
Hi John
It does not talk about intron length as a relevant issue. It talks about protein ratio’s being important for proper alternative splicing.
The question is what are the knobs that control protein ratio’s. One speculation is intron length.
Because a crustacean is different than a human and uses its chromatin differently than a human.
The difference in genome size is probably an indication of how differently the chromatin is used. Do you have a problem with that?
Want more details? Probably not until we have a crustacean ENCODE project for starters. But you seem to “know” it’s junk without definitive data.
But here is one speculation for starters just between vertebrates and invertebrates:
http://www.isj.unimo.it/articoli/ISJ361.pdf
Oh by golly, the DNA could be used differently between vertebrates and invertebrates because they use their epigenomes for different purposes, and a different epigenome requires a different genome.
To quote Gomer Pyle, surprise, surprise, surprise.
Just to repeat, if organisms use their epigenomes differently, there genomes require corresponding size differences. The size of the genome isn’t just about how many genes are needed, but how much material is needed to support the epigenome. You refuse to accept the implications of the article I provided:
http://science.sciencemag.org/content/338/6108/758
PS
How’s your junkDNA camp going to hustle for grants with such a boring attitude that DNA is mostly junk? You think phylogenetic reconstructions are sexy enough projects to inspire the US taxpayers to pony up the sort of big bucks going to ENCODE? How about joining the mainstream and join the ENCODE bandwagon. What is there to lose?
There are a lot of mechanisms that control protein ratios. There are often feedback mechanisms.
Sometimes microRNAs are used gene regulation where maintaining ratios is important.
Here is a suggested feedback mechanism regarding beta catenin. It may not be the answer you were looking for, but it’s what my instincts suspect is an OK explanation so far.
stcordova,
Hi Sal
You have identified perhaps the most important diagram in cancer research. Beta catenin controls the cell cycle and some amount of alternative splicing. Vitamin d up regulates DKK which down regulates beta catenin. This is why controlling blood vitamin d levels is such a powerful cancer prevention strategy. The proteins that beta catenin transcribes can create cancer stem cells. The proteins APC etc are what you call a destruction mechanism which takes beta catenin and breaks it down to its amino acid components. This is how the ratio of beta catenin is controlled in the cell. The protein that controls cellular oxygen levels also has a similar destruction mechanism. I was aware that micro RNA’s are also controlling ratios. My interest is whether intron lengths were possible a part of the timing mechanisms. Great diagram, thanks again for the feedback.
Whose speculation is that, and is there a basis for it? The paper says that protein ratios are in this case controlled by the Wnt pathway, and that has nothing to do with intron length as far as I can see. That would seem to falsify the speculation.
Sal, Can you explain why scientific research funding should be guided by what the U.S. public can be persuaded to find “sexy”?
colewd,
As you’ve probably gathered, unless there is intense interest to conduct a laboratory experiment, it is hard to establish important questions like you’ve posed.
Now in general there is this hypothesis, which you may have encountered:
http://genesdev.cshlp.org/content/22/17/2342.long
And as you gathered, a lot of claims are probably an incomplete picture of what goes on. I’ve not gotten any hits on beta catenin regulation, doesn’t mean introns don’t have a role.
As far as why microRNAs have receive attention. In some respects they are ideal therapeutic drug targets. Hence, I hold a little caution about enthusiasm for microRNAs being the sole explanation for regulation, not to mention it is a known fact microRNAs are usually part of a redundant alternative regulatory network which provides for continued operation in case of failure of another regulatory network.
There has unfortunately not be a lot of interest (as far as I can tell) in targeting intron behavior as a therapeutic target.
Nice to hear from you, and thanks again for raising awareness about vitamin D, and it reminded me about my encounter with my doctor which I had neglected until recently.
John Harshman,
The paper Sal provided investigates Intron length in more detail. Here is a paper that I had heard about which is referenced in Sal’s paper. Why would a transcription take 16hours? Why would an intron be that long?
oogle Scholar
↵ Tennyson, C.N., Klamut, H.J., Worton, R.G.(1995) The human dystrophin gene requires 16 hours to be transcribed and is cotranscriptionally spliced. Nat. Genet.
Here’s a hypothesis: It takes 16 hours because it just happened and there is no selection in that particular gene against it, not because it’s regulated just for that purpose. A shorter gene would be as good, but perhaps a longer one would not. If that hypothesis is true, the introns are still junk, even if bulk DNA is considered functional.