I think if you asked Itskov about that, he'd probably say something along the lines of "given enough time to live, people get other priorities besides being assholes." He's a reformed Russian oligarch, for fnord's sake; it's hard to get more proof-of-concepty than that.
...that being said, it certainly would slow down social change.
But, hey, his timeline includes Surrogates in two years. Probably not something that'll really happen.
Well, yes... but I did said medicine. I'm pretty sure that falls under "other reasons.":)
For what it's worth, I'm not really a medical person at heart, even though I'm surrounded by medical stuff constantly. I'd rather navel-gaze at the LUCA or something. But, yeah, comparative genomics is critical in a wide range of natural resources industries; fisheries and forestry in particular are very concerned about defending their stocks from diseases. (At least, that's what the grant in my job description said.)
My lab works on protozoan parasite genomes in the context of gene network evolution. At a hospital. I'm a doctoral student in computational biology, and my core expertise is sequence analysis. I promise you that I'm well aware of what I said.
I think you, honestly, misread. My point was that sequencing random organisms is not medically useful; it's focusing on diseases (to divine means of attack) or some carefully-selected model organism (to understand a simplified version of ourselves) that brings us important information. And for the most part, we've got the data on the useful models for humans. No one is rushing out and proposing that we replace S. cerevisiae with some obscure protozoan for studying the cell cycle—that work is done.
Health research is much more more now concerned with studying the subtleties of the human genome itself, such as with the ENCODE project, and comparisons with other species have shown us that the new information we've garnered from ENCODE—the huge portion of the genome that's transcribed—is not very conserved at the sequence level between species. I think it's very unlikely that the cure for, say, autism, will involve any cross-species comparisons. Of course, there will always be viruses and pathogenic bacteria, as well as fungal, protozoan, and metazoan parasites.
While certainly there's a lot to do for studying such diseases and their models in depth, those aren't genomes that we're particularly concerned with comparing back to humans, except perhaps to make sure that we can identify a unique drug target that won't harm the patient. That falls under the "evolution of pathogens" part, although perhaps "predicting" was the wrong word.
So, ultimately, my point is: genome comparisons and evolutionary biology are both absolutely essential to medicine, but not all of them. There are a lot of species out there that we're sequencing now which will never be of medical value, and a lot of stuff we're learning about evolution isn't directly applicable either. No human lives will ever hinge on knowing the composition of the Pelagibacter ubique genome, although the environmental genomics knowledge that resulted has certainly opened our eyes to the importance and complexity of the microbial communities that live in symbiosis with the human body. It doesn't really matter that the extracellular matrix is primarily made up of re-used protein domains that are over a billion years old, although ECM research will one day cure a wide variety of human ailments.
Without a doubt that's true, and I'd never propose to debate it—it seems sometimes I spend half my life staring at NCBI and ENSEMBL records; to be honest, I can't think of a corner of the biological sciences that hasn't been affected in some way by the techniques or results of the HGP. That being said, I felt that it was important to ward off flayzernax's discussion of the topic, particularly since it was starting to accrue some supremely lazy mod points.
Except that my first take on this was what innovations.
Every single drug and test invented in the past twelve years has been absolutely dependent on understanding the sequence of the human genome and how its components play together. 15 new anti-cancer drugs were approved in 2012 alone, the most bountiful year for FDA approvals since 1995.
Yes, a lot of stuff is hyped up, but I've seen precious little utility for genomics in the trenches.
Then you've never heard of this? Or this? How much more trench-y do you need?
I think this study is total BS - especially since the methodology is hidden.
The third page links to the full report, noting that it includes the full methodology behind the study. The word "methodology" is right there in bright blue.
Sequencing the human genome is interesting, but the real key is sequencing other genomes and comparing them. That's happening with abandon now.
As far as human health is concerned, the primary questions are (a) how do we work? (b) how do the things that interact with us work? and (c) how does our environment affect us? While many model organisms provide excellent snapshots of simpler genomes that we can use to unravel complex mechanisms (like cell division in yeast), comparative genomics really only teaches us about evolution. It's not relevant to medicine, outside of predicting the evolution of pathogens. We're not benefiting human medicine by sequencing, say, red pandas or sea turtles, although these things are certainly important for other reasons. There are occasionally exceptional genomes, like the naked mole rat (immune to cancer), but these are rare.
And we're finally getting a handle on what controls the genome and how all the little pieces fit together.
The biggest recent contributor to that has been ENCODE, which, again, was a direct analysis of human data and did not involve any other species.
That should yield some better therapies but aside from a few edge cases in cancer treatment, there isn't much out there. And it's not like these cancer treatments have overwhelmingly improved survival - improvements of 20 - 50% are typical. Nothing to sneeze at, but not the Holy Grail.
We picked all the low-hanging fruit like phenylketonuria as soon as it became technologically feasible. Problems like cancer and severe autism are extremely complex, and the only hope we have to tackle them is through an extremely intimate understandinf of the human genome.
(Or, less in line with the joke, understanding of evolution and our place in it. I guess that's sort of like peace, but the Judean People's Front won't have any of it.)
Then look at the report—there're plenty of more specific measures. Six billion dollars in federal and state tax revenue, 293 billion dollars in paid salaries, and 277,000 highly-skilled jobs created or supported. The trillion-dollar figure isn't super sensical at first glance, but there are somewhat more meaningful figures in there.
In the De Menezes case, the police claims were challenged by Underground staff. Perhaps they'll still get away with quite a lot of misconduct, but every lie they have to construct to do so increases the chance that they will be seen through.
Actually this is a perfect example of how claims of unavailable evidence can help convict police. London Underground reported that the cameras were definitely functional, thereby helping to expose the cover-up. If there had been no CCTV cameras in the station, it's probable the final charges wouldn't have been as high as they were.
I think it still leaves room for doubt, doubt that previously was much harder to place. If we assume the police are deliberately using (for example) scheduled maintenance windows to commit brutality, and the suspect is not aware the camera is disabled, then due to Bayesian reasoning we can say with certainty that the officer is more likely to be lying even when there's a legitimate technical failure. (Although we have no way of knowing how much more likely without a lot of data that has not yet been made and, anyway, wouldn't be obtainable.)
1. Cop knows the camera is working + suspect expects the camera to be working -> brutality claims easily proven/disproven, both parties have a disincentive to claim brutality occurred
2. Cop knows the camera isn't working + suspect expects the camera to be working -> brutality can't easily be proven, cop knows there won't be any evidence
3. Cop doesn't know the camera isn't working -> as #1
Obviously there are other factors at work like judgement of character, but the mere fact that the officer would be more confident in being able to get away with brutality should make even legitimate reasons cause a heightened suspicion.
It's still a step in the right direction—it's no longer the police's word vs. the suspect's, but "the police officer says he was having convenient technical difficulties at the same time his account of the incident is in conflict with the suspect's." It looks worse in court, since police will be more than happy to produce video when they are innocent. This is much better than no camera at all.
But on that topic, what's the deal with this company? Their staff roster doesn't seem to include any actual mathematicians or statisticians. They don't mention anywhere in the press release (or any other articles I could find on their site) how they arrive at their figures. It's just Jobs-style gut feelings and marketing experience. Even if their predictions are right it's impossible to trust such unbacked assertions... and, ultimately, it's irrelevant to the real players: Microsoft, Google, and Apple all employ scores of PhDs and industry veterans who could forecast circles around them.
At the end of the video, the robot arm folds up; the way it did this and the framing of the shot reminded me of WALLE in such a way that I thought it was about to roll off and make another copy of the artwork it had just constructed, like a demented robot gardener—alas, it had no wheels.
Note to self: when rich, get robotic arm to sculpt perlin volume topiaries.
No, the compile worked fine; the problem only existed in the imagination of Eclipse's code analysis (syntax checking.) std::size_t as defined in stddef.h was conflicting with size_t in c++config.h when using namespace std.
Slashdot Mirror
Right; now show me a fully-actuated android doing the same thing. Wee bit trickier, yes?
Maybe, maybe not. But he'd sure say that!
Mesmerisingly, no matter how long I spend on a post here, there always seems to be some way to make them clearer in hindsight.
I think if you asked Itskov about that, he'd probably say something along the lines of "given enough time to live, people get other priorities besides being assholes." He's a reformed Russian oligarch, for fnord's sake; it's hard to get more proof-of-concepty than that.
...that being said, it certainly would slow down social change.
But, hey, his timeline includes Surrogates in two years. Probably not something that'll really happen.
Well, yes... but I did said medicine. I'm pretty sure that falls under "other reasons." :)
For what it's worth, I'm not really a medical person at heart, even though I'm surrounded by medical stuff constantly. I'd rather navel-gaze at the LUCA or something. But, yeah, comparative genomics is critical in a wide range of natural resources industries; fisheries and forestry in particular are very concerned about defending their stocks from diseases. (At least, that's what the grant in my job description said.)
My lab works on protozoan parasite genomes in the context of gene network evolution. At a hospital. I'm a doctoral student in computational biology, and my core expertise is sequence analysis. I promise you that I'm well aware of what I said.
I think you, honestly, misread. My point was that sequencing random organisms is not medically useful; it's focusing on diseases (to divine means of attack) or some carefully-selected model organism (to understand a simplified version of ourselves) that brings us important information. And for the most part, we've got the data on the useful models for humans. No one is rushing out and proposing that we replace S. cerevisiae with some obscure protozoan for studying the cell cycle—that work is done.
Health research is much more more now concerned with studying the subtleties of the human genome itself, such as with the ENCODE project, and comparisons with other species have shown us that the new information we've garnered from ENCODE—the huge portion of the genome that's transcribed—is not very conserved at the sequence level between species. I think it's very unlikely that the cure for, say, autism, will involve any cross-species comparisons. Of course, there will always be viruses and pathogenic bacteria, as well as fungal, protozoan, and metazoan parasites.
While certainly there's a lot to do for studying such diseases and their models in depth, those aren't genomes that we're particularly concerned with comparing back to humans, except perhaps to make sure that we can identify a unique drug target that won't harm the patient. That falls under the "evolution of pathogens" part, although perhaps "predicting" was the wrong word.
So, ultimately, my point is: genome comparisons and evolutionary biology are both absolutely essential to medicine, but not all of them. There are a lot of species out there that we're sequencing now which will never be of medical value, and a lot of stuff we're learning about evolution isn't directly applicable either. No human lives will ever hinge on knowing the composition of the Pelagibacter ubique genome, although the environmental genomics knowledge that resulted has certainly opened our eyes to the importance and complexity of the microbial communities that live in symbiosis with the human body. It doesn't really matter that the extracellular matrix is primarily made up of re-used protein domains that are over a billion years old, although ECM research will one day cure a wide variety of human ailments.
Without a doubt that's true, and I'd never propose to debate it—it seems sometimes I spend half my life staring at NCBI and ENSEMBL records; to be honest, I can't think of a corner of the biological sciences that hasn't been affected in some way by the techniques or results of the HGP. That being said, I felt that it was important to ward off flayzernax's discussion of the topic, particularly since it was starting to accrue some supremely lazy mod points.
But who could reject such a sincere message of honesty and dedication?!
Woah, woah, what?
Except that my first take on this was what innovations.
Every single drug and test invented in the past twelve years has been absolutely dependent on understanding the sequence of the human genome and how its components play together. 15 new anti-cancer drugs were approved in 2012 alone, the most bountiful year for FDA approvals since 1995.
Yes, a lot of stuff is hyped up, but I've seen precious little utility for genomics in the trenches.
Then you've never heard of this? Or this? How much more trench-y do you need?
I think this study is total BS - especially since the methodology is hidden.
The third page links to the full report, noting that it includes the full methodology behind the study. The word "methodology" is right there in bright blue.
Sequencing the human genome is interesting, but the real key is sequencing other genomes and comparing them. That's happening with abandon now.
As far as human health is concerned, the primary questions are (a) how do we work? (b) how do the things that interact with us work? and (c) how does our environment affect us? While many model organisms provide excellent snapshots of simpler genomes that we can use to unravel complex mechanisms (like cell division in yeast), comparative genomics really only teaches us about evolution. It's not relevant to medicine, outside of predicting the evolution of pathogens. We're not benefiting human medicine by sequencing, say, red pandas or sea turtles, although these things are certainly important for other reasons. There are occasionally exceptional genomes, like the naked mole rat (immune to cancer), but these are rare.
And we're finally getting a handle on what controls the genome and how all the little pieces fit together.
The biggest recent contributor to that has been ENCODE, which, again, was a direct analysis of human data and did not involve any other species.
That should yield some better therapies but aside from a few edge cases in cancer treatment, there isn't much out there. And it's not like these cancer treatments have overwhelmingly improved survival - improvements of 20 - 50% are typical. Nothing to sneeze at, but not the Holy Grail.
We picked all the low-hanging fruit like phenylketonuria as soon as it became technologically feasible. Problems like cancer and severe autism are extremely complex, and the only hope we have to tackle them is through an extremely intimate understandinf of the human genome.
Here you go: 01101000 01110100 01110100 01110000 00111010 00101111 00101111 01110111 01110111 01110111 00101110 01111001 01101111 01110101 01110100 01110101 01100010 01100101 00101110 01100011 01101111 01101101 00101111 01110111 01100001 01110100 01100011 01101000 00111111 01110110 00111101 01000100 01000101 00101101 00111000 01111001 01001111 00110011 01100110 01101110 01001010 00110100
Brought peace?
(Or, less in line with the joke, understanding of evolution and our place in it. I guess that's sort of like peace, but the Judean People's Front won't have any of it.)
Then look at the report—there're plenty of more specific measures. Six billion dollars in federal and state tax revenue, 293 billion dollars in paid salaries, and 277,000 highly-skilled jobs created or supported. The trillion-dollar figure isn't super sensical at first glance, but there are somewhat more meaningful figures in there.
This is about the human genome, you goof, not crops. The money is from medical innovations. RTFA!
No, it's more like "please don't stop funding science, specifically this one part."
In the De Menezes case, the police claims were challenged by Underground staff. Perhaps they'll still get away with quite a lot of misconduct, but every lie they have to construct to do so increases the chance that they will be seen through.
Actually this is a perfect example of how claims of unavailable evidence can help convict police. London Underground reported that the cameras were definitely functional, thereby helping to expose the cover-up. If there had been no CCTV cameras in the station, it's probable the final charges wouldn't have been as high as they were.
Oh, I did. Of course, best interests need to be compromised to serve the needs of the people.
Well, yeah—it's always in your best interest to have a recording of your own actions. At least, as long as you can't be forced to give it up.
I think it still leaves room for doubt, doubt that previously was much harder to place. If we assume the police are deliberately using (for example) scheduled maintenance windows to commit brutality, and the suspect is not aware the camera is disabled, then due to Bayesian reasoning we can say with certainty that the officer is more likely to be lying even when there's a legitimate technical failure. (Although we have no way of knowing how much more likely without a lot of data that has not yet been made and, anyway, wouldn't be obtainable.)
Obviously there are other factors at work like judgement of character, but the mere fact that the officer would be more confident in being able to get away with brutality should make even legitimate reasons cause a heightened suspicion.
It's still a step in the right direction—it's no longer the police's word vs. the suspect's, but "the police officer says he was having convenient technical difficulties at the same time his account of the incident is in conflict with the suspect's." It looks worse in court, since police will be more than happy to produce video when they are innocent. This is much better than no camera at all.
But on that topic, what's the deal with this company? Their staff roster doesn't seem to include any actual mathematicians or statisticians. They don't mention anywhere in the press release (or any other articles I could find on their site) how they arrive at their figures. It's just Jobs-style gut feelings and marketing experience. Even if their predictions are right it's impossible to trust such unbacked assertions... and, ultimately, it's irrelevant to the real players: Microsoft, Google, and Apple all employ scores of PhDs and industry veterans who could forecast circles around them.
At the end of the video, the robot arm folds up; the way it did this and the framing of the shot reminded me of WALLE in such a way that I thought it was about to roll off and make another copy of the artwork it had just constructed, like a demented robot gardener—alas, it had no wheels.
Note to self: when rich, get robotic arm to sculpt perlin volume topiaries.
I was looking at it earlier; I'll keep it in mind.
That's where things got terrifying. Strawberry's CPAN wanted MSYS.
No, the compile worked fine; the problem only existed in the imagination of Eclipse's code analysis (syntax checking.) std::size_t as defined in stddef.h was conflicting with size_t in c++config.h when using namespace std.