9 ms·
Is Most of Our DNA Garbage?
- damon_c 12y agoWhen you open a binary file in a text editor it looks like garbage.
- iLoch 12y agoDoesn't mean all the code that was used to compile that binary is used though.
- msandford 12y agoI heard about this "98% of DNA is junk" claim a while ago. http://www.livescience.com/31939-junk-dna-mystery-solved.html http://www.livescience.com/31939-junk-dna-mystery-solved.htm... Right, because a single cell transforms into a whole animal (horse, giraffe, whatever) and that takes zero information. Really? Somehow I doubt it. A multicellular organism is a really big, complicated state machine and the "junk" DNA is what's needed in order to go from a single cell to many billions of cells organized in the particular fashion that they are. To suppose that everything that doesn't directly code for a protein is "junk" is incredibly arrogant; just because you don't currently understand something doesn't mean that it cannot be. As an engineer I'm guilty of this all the time; something I don't know how to do is impossible but once someone explains to me how to do it, it's trivial. I've had this happen enough times that I can now recognize that my "impossible" isn't necessarily actually impossible, just that it might be impossible to me right now.
- jhallenworld 12y agoDon't underestimate the creative capacity of small (to encode) algorithms. Perhaps development uses something like simulated annealing to create structures which match required properties. If it works like this, an exact template is not required to be encoded in the genes. In other words, the ultimate source of the design is thermal randomness, but guided by constraints encoded in the genes.
- msandford 12y agoWell we're starting to see Turing vindicated on the way that a lot of biological stuff happens. http://genomebiology.com/2013/14/1/101 http://genomebiology.com/2013/14/1/101 Sure it doesn't have an exact template of _precisely_ how to do everything, but there's a lot of information there. Maybe it's not huge relative to the amount of theoretical information DNA contains, but I strongly suspect that the 98% is junk claim will eventually be found to be laughably wrong.
- collyw 12y agoAt the same time Pine tree genomes are 20 times the size of our own, but have a similar number of coding genes. Do we really think all of the extra DNA is needed?
- taco_emoji 12y agoBinary files didn't evolve.
- Crito 12y agoThe current state of things seems more like opening ASCII source code in a hex editor to me. We can sequence DNA, but that raw data isn't something that we can mentally digest very easily at all.
- fuligo 12y agoAs programmers we're in a unqiue position to get a sense of this, actually. It's very very old spaghetti code. Some of it essential, there are even some cleverly re-used common routines, some of it is just ballast, some of it can't be taken out but isn't getting executed either, some of it now serves a different purpose, and none of it comes with annotations that allow us to easily find out which is which.
- mod 12y agoVery cool analogy. At least we're getting to the point where we can flip switches and monitor for changes. Debug print!
- thaumasiotes 12y agoWhy do you think it's an analogy?
- benihana 12y agoBecause our DNA isn't a literal computer program written by humans in the past 50 years.
- thaumasiotes 12y agoThat comparison was never drawn, though. It was just observed that DNA is old (true in fact) spaghetti code (true in fact) with clever code reuse (true in fact), dead weight (conjectured but plausible), stuff with no apparent purpose that nevertheless can't be removed without destroying functionality (true in fact), stuff that used to serve one purpose but currently serves a completely different purpose (true in fact)... all of these phenomena can be directly observed in a working cell. Granted, "spaghetti code" is a value judgment, but I challenge you to find anyone who thinks it doesn't apply to DNA.
- charonn0 12y agoThis is exactly how I like to think of DNA: a twisted mass of legacy code. Rife with Heisenbugs, the slightest change can have profound side-effects in seemingly unrelated subsystems. Even sections that are literal "junk" must be retained since everything uses GOTOs with hard-coded line numbers.
- MollyR 12y agoWhile there might be some signal noise ratio, I highly doubt we have any junk dna. I imagine a good modern analogy here would be http://en.wikipedia.org/wiki/Evolvable_hardware http://en.wikipedia.org/wiki/Evolvable_hardware. http://www.damninteresting.com/on-the-origin-of-circuits/ http://www.damninteresting.com/on-the-origin-of-circuits/ in the two articles I mentioned,I think the way the "offspring" circuit worked in an extremely baffling(complex) way relying in physics in way beyond human design (aka the way the disconnected logic gates were still necessary for functionality) have insight into some of possible complex interactions of dna. They look like unnecessary garbage until you try to remove them and break a ton of stuff.
- NathanKP 12y agoAccording to the article: On average, each baby is born with roughly 100 new mutations. If every piece of the genome were essential, then many of those mutations would lead to significant birth defects, with the defects only multiplying over the course of generations; in less than a century, the species would become extinct. The point this article is making is that new techniques are finding that some portions of DNA assumed to be junk might have a use, and removing or changing them certainly does have an effect. But that doesn't mean we should jump ahead too far and say that none of the DNA is junk, because there are obviously some things that have no effect when changed.
- simcop2387 12y agoI wouldn't be surprised if some of it really has no effect and that's the whole point. As you mentioned if there's 100 new mutations per baby then if the coding is as dense as possible that means each mutation will cause some significant change. If it's got a lot of coding that doesn't do anything, by probabilities it'll likely not change anything important and cause no ill effects. Having that extra unused DNA could pose an evolutionary advantage in the face of an imperfect copying mechanism.
- ajuc 12y agoAssuming mutation probabiity don't depend on the length of the copied DNA - copying 40 MB you will probably introduce 10 times more mutations than if you just copied 4 MB. only 10% of them will on average change the important code, so the end result is the same. So I don't see how NOP padding helps.
- jhallenworld 12y agoI thought the introns were not well preserved between generations and that this is strong evidence that they really are junk.
- Animats 12y agoFlowering plants have huge amounts of DNA, 1-2 orders of magnitude more than mammals. It's like they haven't evolved subroutines or macros or something comparable.
- azeirah 12y agoI'm excited to see the headline "Nature's usage of macros in insect DNA" appear on hacker news next week.
- mbreese 12y agoYou might get a kick out of insect patterning based upon morphogen protein concentration gradients then... http://en.wikipedia.org/wiki/Drosophila_embryogenesis http://en.wikipedia.org/wiki/Drosophila_embryogenesis
- damon_c 12y agoThe idea that nature may have evolved coding design patterns to make "cleaner more efficient" code is amazing. Are flowering plants just written in a language that doesn't have "for loops" so everything is just written out line by line?
- acqq 12y agoNo. The compactness of the DNA "program" is an evolutionary trade off, just like in the engineering. For some species to have a significantly more compact code now there should have existed some evolutionary benefit among the ancestors for that. http://en.wikipedia.org/wiki/Genome_size#Genome_reduction_in_obligate_endosymbiotic_species http://en.wikipedia.org/wiki/Genome_size#Genome_reduction_in...
- damon_c 12y agoInteresting, and from the same page, Drake's rule says that similarly to programming, the larger the "codebase", the slower the mutation (new feature launch) rate. http://en.wikipedia.org/wiki/Genome_size#Drake.27s_rule http://en.wikipedia.org/wiki/Genome_size#Drake.27s_rule
- jonchang 12y agoThere's an interesting recent manuscript that subdivides the obsolete "junk DNA" classification into things like "garbage DNA", "rubbish DNA", "Lazarus DNA", "zombie DNA", etc. http://gbe.oxfordjournals.org/content/early/2015/01/28/gbe.evv021.short http://gbe.oxfordjournals.org/content/early/2015/01/28/gbe.e...
- OldSchoolJohnny 12y agoPerhaps when the right conditions arise the "junk" turns out to be useful after all?
- jostmey 12y agoCouldn't agree more. Evolution as a process has no regard for simplicity or elegance. The solutions that Natural Selection "comes" up with are random. I guess you could say that all organisms have a design, but that the design is irrational. A lot of people expect that at the molecular level life is organized and sensible. But why should that be the case? On the macroscopic level, living organisms are a mess. It is my belief that at the biochemical level the chemical reactions are as organized as the branches on a tree---that the chaotic design of life at the microscopic level mirrors its design at the macroscopic level.
- api 12y agoIt's best to throw out all these value judgements. Evolution isn't purely random. There's nothing random about what works and what doesn't. Over time, differential selection and other effects (e.g. sexual selection) work to transfer information about the environment into the genome. Evolution is a learning process. It's just that the representation that it generates is difficult for us linear thinkers to interpret. It only appears irrational from an anthropocentric point of view. It's not rational to us because we, having brains that work in a certain way, are biased toward seeing things as linear chains of cause and effect. That's how we think and that's how we like to build stuff, but obviously that's not the only way stuff can be built. For all we know elsewhere in the universe there are beings that think in massively parallel super-holistic causality-matrix terms. To them evolution's designs would seem perfectly rational, while a UML block diagram would seem insane. "Nothing has only one cause or one effect," they would mutter... in a language in which every word implies everything to varying degrees and each syllable of a sentence must be parsed in parallel with all others.
- sumitviii 12y agoI think parent meant that new mutations are random. Or at least look random.
- plikan13 12y ago> Couldn't agree more. Evolution as a process has no regard for simplicity or elegance. The solutions that Natural Selection "comes" up with are random. I guess you could say that all organisms have a design, but that the design is irrational. Except of course that evolution comes up with solutions that are so clever and so intricate that we, with all our scientific brains, have barely scratched the surface of understanding it. Now if it acts intelligently, walks intelligently and quacks intelligently, then maybe we should just call it intelligent?
- coderzach 12y agoThis reminds me of a C program I once saw. Certain functions in the code were dependent on the stack order. If you moved any thing around the (10mm+ LoC) app would fail to compile. Therefore any time anyone wanted to change these functions they would just append more code to the bottom, never touching any of the existing code. Most of it was garbage, but if you touched it at all, everything would break.
- nashashmi 12y agoI am surprised at the volume of agnostic comments here. Before I begin, let me explain my position. I believe in God, and believe in evolution as coming from God himself. Second, I distrust any behavior of arrogance or pride that comes from a I-know-it-all person. Third, the very people who often do create, or innovate, or discover are far more humble and more curious than the previous people I described. Having said that, there is some fanboy culture and attachment with science and DNA tinkering. Some "discoveries" and claims sometimes come from such fanboys. People on HN are not different. In fact majority of the comments here signal as coming from fanboys. Now to the topic of DNA: DNA, if a product of evolution, could not be mostly composed of garbage because nature has its own "garbage collector." Much of the DNA is important and significant in some shape or form. It won't be apparent in a simple experiment, but it may be apparent over a lifetime or maybe over generations. There is evidence to back this up: a recent NPR show, i forget what, claimed that your grandfather's hunger at the age of 9 influenced your chances of heart attack. This brings up DNA's influence across lifetimes. And the complications are so much deeper. Our fight to try to understand DNA is incredible. We are not there completely. And maybe this is the longtail part of science. But we will understand it more and more, and imagine the information we extract then. I bet much of it will debunk everything we so arrogantly claim today.
- taco_emoji 12y ago> nature has its own "garbage collector." Ah, no, it doesn't. If you think "survival of the fittest" will somehow magically cull non-functional, non-deleterious DNA out of the genome (except by accident), then you don't understand evolution. It's kind of hypocritical to call others "arrogant" when you're opining on a topic you clearly don't grasp yourself.
- nashashmi 12y agoNope. Not being hypocrite here. DNA that is not needed will wither away. And I do understand evolution. Quite well actually simply because I have debated it far more rationally than those who have outright rejected it and those who have unquestionably accepted it. And there happen to be a lot few mysteries with my explanation. Think of DNA as computer code. Think of each creature as a robot with computer code built in. Now think of DNA as also having the code to recreate the creature. Overloaded DNA will collapse. Or incredibly sophisticated biology will be able to handle overloaded DNA. Either way, too much complexity breaks. You may call it "Survival of the fittest" implying chaos theory but I call it harmonious morphology implying benign force.
- finnh 12y agoDidn't they recently realize that huge swaths of "junk" DNA are actually use to regulate RNA expression? So, uh, not junk? Not the original article I'm thinking of, but related: https://www.uam.es/personal_pdi/ciencias/genhum/bibliogenoma/RNAregulationnewgenetics.pdf https://www.uam.es/personal_pdi/ciencias/genhum/bibliogenoma...
- jostmey 12y agoYou state it as fact but it remains a matter of heated debate. If all the DNA were important then why are large swaths of DNA allowed to undergo genetic drift?
- collyw 12y agoWhy then do pine trees with a similar number of genes to our own, have nearly 20 times larger genome? Are you assuming they are more regulated?
- rogerbinns 12y agoI highly recommend reading "DNA seen through the eyes of a coder" at http://ds9a.nl/amazing-dna/ http://ds9a.nl/amazing-dna/ It is a very nice explanation, shows many similarities, while also making Intercal seem sane.
- disjointrevelry 12y agoThe way I see it is DNA is much like a very long running program meant to run for millions to billions of years. Our current life is an instance of some code in the DNA. Due to the longevity of the purpose of DNA, and our short span in an instance, it is very, very difficult to ascertain what some parts may be about, or what it's for. Not to mention the high compressibility of information in DNA, some aspects might never really reveal themselves unless met with an instance in an environment it is "meant to run in".