Another stupid piece of DNA worship 9 Feb 2008 By Matt Ridley, in Time: … by the end of this century, if not sooner, biotechnology may have reached the point where it can take just about any DNA recipe and read off a passable 3-D interpretation of the animal it would create. So long as you also know the developmental machinery, the necessary ecological conditions, the structure of the cells, the maternal investment involved… in other words, if you know the facts about the structure and biology of the organism, you’ll be able to read off the structure and biology of the organism, just from the DNA. Err…. Evolution
Evolution Evolution quotes: Socialism 9 May 20129 May 2012 To the biologist the problem of socialism appears largely as a problem of size. The extreme socialists desire to run every nation as a single business concern. I do not suppose that Henry Ford would find much difficulty in running Andorra or Luxembourg on a socialistic basis. He has already… Read More
Biology Darwin was not badly received by the church 26 Nov 2009 Robert J. Berry is a geneticist at University College London. He is also an evangelical Christian and has written a number of works on the compatibility of religion (his kind, anyway) and evolution. He has a quite accurate letter in today’s Nature. Since that is behind a paywall, I have… Read More
Evolution Browsing through the Philosophical Transactions on species and generation 30 Aug 2008 One of the major events in the history of science was the foundation of a number of published communications, so that the results of observation and research could be relatively quickly shared amongst scholars, and one of the first of these was the Philosophical Transactions of the Royal Society of… Read More
Maybe, or maybe not… “Computers in the future may weigh no more than 1.5 tons.” Popular Mechanics, forecasting the relentless march of science, 1949 “I think there is a world market for maybe five computers.” Thomas Watson, chairman of IBM, 1943 “There is no reason anyone would want a computer in their home.” Ken Olson, president, chairman and founder of Digital Equipment Corp., 1977 “Who the hell wants to hear actors talk?” H.M. Warner, Warner Brothers, 1927. “Heavier-than-air flying machines are impossible.” Lord Kelvin, president, Royal Society, 1895.
Ian – I think you missed the essential point: DNA doesn’t carry enough information to predict a 3D model.
Ian – I think you missed the essential point: DNA doesn’t carry enough information to predict a 3D model.
Ian – I think you missed the essential point: DNA doesn’t carry enough information to predict a 3D model.
Ian – I think you missed the essential point: DNA doesn’t carry enough information to predict a 3D model.
Why, John, this isn’t an argument from incredulity, is it? The comment may be hyperbole, but is it really that far-fetched? If we allow ourselves the assumption that such a DNA recipe would be based on organisms we have around us now (even if not in 90 years), I don’t think it looks so ridiculous. To answer your points: developmental machinery, This is pretty well conserved, so I don’t see it as a big problem. The devil will be in the detail, of course. the necessary ecological conditions, For many organisms, this is not a problem, as they are homeostatic enough to be buffered against the environment. Human children look fairly alike if they’re born in the Australian summer or the Finnish winter. For organisms that aren’t so homeostatic, the genome will give clues about the environment (e.g. thermal tolerance), so that could be guessed at to some extent. Even if we can’t work out the exact 3D structure, we might still pick up the main aspects – pine trees look like pine trees, even if their precise structure can vary. Even if specific environmental cues are needed for development to proceed, they may not have to be known. It may be enough to recognise the protein at the start of a signal transduction pathway, and then work out the effect of that pathway being switched on at the right time. the structure of the cells, Again, I think much could be worked out. Want to know if the cell has a nucleus? Look for the proteins that form a nuclear pore! Want to know what the lipids of the cell membranes are like? Check the lipid synthesis genes. etc. etc. the maternal investment involved I think this is the hardest problem but I think a lot could still be worked out. Maternal effects on development can probably be guessed at (at least initially), from homology. Other effects, such as putting a layer of calcium around the offspring, may not matter so much – it might be enough to work out that the embryo is able to absorb nutrients from its immediate surrounding, without knowing how exactly they are packaged. I’m actually not sure if the structure could be “read off”, but I think it might be possible to work it out. Try a reasonable configuration, see how development proceeds. If it doesn’t work, try tweaking it. etc. etc. Bob
Why, John, this isn’t an argument from incredulity, is it? The comment may be hyperbole, but is it really that far-fetched? If we allow ourselves the assumption that such a DNA recipe would be based on organisms we have around us now (even if not in 90 years), I don’t think it looks so ridiculous. To answer your points: developmental machinery, This is pretty well conserved, so I don’t see it as a big problem. The devil will be in the detail, of course. the necessary ecological conditions, For many organisms, this is not a problem, as they are homeostatic enough to be buffered against the environment. Human children look fairly alike if they’re born in the Australian summer or the Finnish winter. For organisms that aren’t so homeostatic, the genome will give clues about the environment (e.g. thermal tolerance), so that could be guessed at to some extent. Even if we can’t work out the exact 3D structure, we might still pick up the main aspects – pine trees look like pine trees, even if their precise structure can vary. Even if specific environmental cues are needed for development to proceed, they may not have to be known. It may be enough to recognise the protein at the start of a signal transduction pathway, and then work out the effect of that pathway being switched on at the right time. the structure of the cells, Again, I think much could be worked out. Want to know if the cell has a nucleus? Look for the proteins that form a nuclear pore! Want to know what the lipids of the cell membranes are like? Check the lipid synthesis genes. etc. etc. the maternal investment involved I think this is the hardest problem but I think a lot could still be worked out. Maternal effects on development can probably be guessed at (at least initially), from homology. Other effects, such as putting a layer of calcium around the offspring, may not matter so much – it might be enough to work out that the embryo is able to absorb nutrients from its immediate surrounding, without knowing how exactly they are packaged. I’m actually not sure if the structure could be “read off”, but I think it might be possible to work it out. Try a reasonable configuration, see how development proceeds. If it doesn’t work, try tweaking it. etc. etc. Bob
Why, John, this isn’t an argument from incredulity, is it? The comment may be hyperbole, but is it really that far-fetched? If we allow ourselves the assumption that such a DNA recipe would be based on organisms we have around us now (even if not in 90 years), I don’t think it looks so ridiculous. To answer your points: developmental machinery, This is pretty well conserved, so I don’t see it as a big problem. The devil will be in the detail, of course. the necessary ecological conditions, For many organisms, this is not a problem, as they are homeostatic enough to be buffered against the environment. Human children look fairly alike if they’re born in the Australian summer or the Finnish winter. For organisms that aren’t so homeostatic, the genome will give clues about the environment (e.g. thermal tolerance), so that could be guessed at to some extent. Even if we can’t work out the exact 3D structure, we might still pick up the main aspects – pine trees look like pine trees, even if their precise structure can vary. Even if specific environmental cues are needed for development to proceed, they may not have to be known. It may be enough to recognise the protein at the start of a signal transduction pathway, and then work out the effect of that pathway being switched on at the right time. the structure of the cells, Again, I think much could be worked out. Want to know if the cell has a nucleus? Look for the proteins that form a nuclear pore! Want to know what the lipids of the cell membranes are like? Check the lipid synthesis genes. etc. etc. the maternal investment involved I think this is the hardest problem but I think a lot could still be worked out. Maternal effects on development can probably be guessed at (at least initially), from homology. Other effects, such as putting a layer of calcium around the offspring, may not matter so much – it might be enough to work out that the embryo is able to absorb nutrients from its immediate surrounding, without knowing how exactly they are packaged. I’m actually not sure if the structure could be “read off”, but I think it might be possible to work it out. Try a reasonable configuration, see how development proceeds. If it doesn’t work, try tweaking it. etc. etc. Bob
Why, John, this isn’t an argument from incredulity, is it? The comment may be hyperbole, but is it really that far-fetched? If we allow ourselves the assumption that such a DNA recipe would be based on organisms we have around us now (even if not in 90 years), I don’t think it looks so ridiculous. To answer your points: developmental machinery, This is pretty well conserved, so I don’t see it as a big problem. The devil will be in the detail, of course. the necessary ecological conditions, For many organisms, this is not a problem, as they are homeostatic enough to be buffered against the environment. Human children look fairly alike if they’re born in the Australian summer or the Finnish winter. For organisms that aren’t so homeostatic, the genome will give clues about the environment (e.g. thermal tolerance), so that could be guessed at to some extent. Even if we can’t work out the exact 3D structure, we might still pick up the main aspects – pine trees look like pine trees, even if their precise structure can vary. Even if specific environmental cues are needed for development to proceed, they may not have to be known. It may be enough to recognise the protein at the start of a signal transduction pathway, and then work out the effect of that pathway being switched on at the right time. the structure of the cells, Again, I think much could be worked out. Want to know if the cell has a nucleus? Look for the proteins that form a nuclear pore! Want to know what the lipids of the cell membranes are like? Check the lipid synthesis genes. etc. etc. the maternal investment involved I think this is the hardest problem but I think a lot could still be worked out. Maternal effects on development can probably be guessed at (at least initially), from homology. Other effects, such as putting a layer of calcium around the offspring, may not matter so much – it might be enough to work out that the embryo is able to absorb nutrients from its immediate surrounding, without knowing how exactly they are packaged. I’m actually not sure if the structure could be “read off”, but I think it might be possible to work it out. Try a reasonable configuration, see how development proceeds. If it doesn’t work, try tweaking it. etc. etc. Bob
I don’t see a problem with this prediction. By that time we will not be starting off from scratch – we will have many model organisms sequenced to base our predictions and no doubt as mentioned, we will have many experiments done to test the hypothesis. There is so much modularity involved that we do not need to design single proteins or pathways from scratch, just learn how to regulate them and fit them together, probably in similar ways to that already done in current organisms.
In a way it’s already possible. Just look up the sequence, what animal it belongs to and look up a picture of that animal. All we’d need is a database of 3D imaged animals. But seriously, I think that we’d be lucky if we could predict the 3D structure of a protein from it’s gene sequence by the end of the century.
Bob, the problem is that the “specification” lies not in the DNA, but in the DNA plus all the other aspects of the organism. Sure we can make inductive inferences based on prior knowledge plus phylogenetics but that is very far from suggesting that the 3-D structure of an organism is somehow encoded in the DNA. As Flaky says, we haven’t even solved the protein folding problem yet. What possible reason, apart from DNA centrism, is there for thinking that the DNA specifies morphology on its own? And yeah, epigenetics, mfs…
Lassi Hippel?inen wrote: Ian – I think you missed the essential point: DNA doesn’t carry enough information to predict a 3D model. I got the point, Lassi. I was just trying to make a lighthearted comment about the unreliability of long-range scientific prognostications. I was also trying to think of an analog to DNA in human technology to use as an illustration but there doesn’t appear to be anything. If you selected some computer code at random, would you be able to infer from it the 3-D structure of the machine that ran it, even if it were the schematics for the box, screen and keyboard and all the gubbins that goes inside?
Epigenetics is not a get-out-of-DNA-jail-for-free card. I’d guess that most of the DNA modifications going on actually proceed by *rules encoded in the genome* (like X chromosome inactivation), and actual consistent epigenetic inheritance not encoded in the genome is in the minority. And what about things like maternal effects? To simulate Drosophila development, yes, you’d need to know that there is a gradient of bicoid in the egg before development even starts. But why couldn’t the egg be simulated, in principle, based on the information in the maternal genome?
…but that is very far from suggesting that the 3-D structure of an organism is somehow encoded in the DNA. Ridley isn’t intending this: The rest is as easy as Dolly the sheep: call up a company that can synthesize the genome, stick it into an enucleated ostrich ovum, implant the same in an ostrich and sit back to watch the fun. Which suggests that he knows that a developmental context is needed. The big aspect I can see missing is maternal RNA, but that could either be provided by the ostrich or it worked out by computer (as I was suggesting). As Flaky says, we haven’t even solved the protein folding problem yet. That’s fine, we’ve got 90 years to go. I guess Ridley is more of an optimist than Flaky. Bob
…but that is very far from suggesting that the 3-D structure of an organism is somehow encoded in the DNA. Ridley isn’t intending this: The rest is as easy as Dolly the sheep: call up a company that can synthesize the genome, stick it into an enucleated ostrich ovum, implant the same in an ostrich and sit back to watch the fun. Which suggests that he knows that a developmental context is needed. The big aspect I can see missing is maternal RNA, but that could either be provided by the ostrich or it worked out by computer (as I was suggesting). As Flaky says, we haven’t even solved the protein folding problem yet. That’s fine, we’ve got 90 years to go. I guess Ridley is more of an optimist than Flaky. Bob
…but that is very far from suggesting that the 3-D structure of an organism is somehow encoded in the DNA. Ridley isn’t intending this: The rest is as easy as Dolly the sheep: call up a company that can synthesize the genome, stick it into an enucleated ostrich ovum, implant the same in an ostrich and sit back to watch the fun. Which suggests that he knows that a developmental context is needed. The big aspect I can see missing is maternal RNA, but that could either be provided by the ostrich or it worked out by computer (as I was suggesting). As Flaky says, we haven’t even solved the protein folding problem yet. That’s fine, we’ve got 90 years to go. I guess Ridley is more of an optimist than Flaky. Bob
…but that is very far from suggesting that the 3-D structure of an organism is somehow encoded in the DNA. Ridley isn’t intending this: The rest is as easy as Dolly the sheep: call up a company that can synthesize the genome, stick it into an enucleated ostrich ovum, implant the same in an ostrich and sit back to watch the fun. Which suggests that he knows that a developmental context is needed. The big aspect I can see missing is maternal RNA, but that could either be provided by the ostrich or it worked out by computer (as I was suggesting). As Flaky says, we haven’t even solved the protein folding problem yet. That’s fine, we’ve got 90 years to go. I guess Ridley is more of an optimist than Flaky. Bob
Reminds me of the ST:Enterprise episode where Archer’s stupid mutt widdles on an alien culture’s sacred tree. Archer protests that the aliens had been supplied with the dog’s DNA; they should have been able to work out that dogs like to pee on vertical objects, just for the hell of it. Porteous: that series’ winner of the Wesley Crusher “Please Space That Character” Award.
The main problem that comes to my mind in the assertion that you “can take just about any DNA recipe and read off a passable 3-D interpretation of the animal it would create” is the hormonal/nutritional environment provided by the mother during pregnancy. (I suppose that could be counted as environment.) It presents a kind of chicken-or-the-egg problem: once you have simulated a complete working organism, then you can figure out how its body would affect the offspring, and then predict how that offspring would form.
It brings to mind numerous episodes of Star Trek: The Next Generation in which somebody’s DNA got scrambled, the Doctor, the Ensign, or the Janitor looked at a big picture of a double helix, and fixed the poor crew member who was temporarily a monster. I guess “DNA recipe” is vague enough to include all that’s needed, but even so I doubt such recipes would be available for many organisms, even with accelerated data collection.
I guess “DNA recipe” is vague enough to include all that’s needed, but even so I doubt such recipes would be available for many organisms, even with accelerated data collection. We have hundreds of genomes now, although the sample is phylogenetically biased. I think we can manage a bit more by the end of this century, don’t you?