Also, for two days and the first one thousand orders, Veritas -- a separate company from the above -- is offering a 30X full genome sequencing for $199. Even with the costs of biotech plummeting, that is a great deal.
In my estimation, this is a big deal. If the human race can round up enough volunteers to both get their genomes sequenced and allow researchers to study those sequences, they can start to do large (million genome plus) genome-wide association studies (GWAS).
We'll be able to start really getting things locked down. It turns out that a lot of the things that we are interested in aren't determined by a few genes of large effect, but by a large number of genes of small effect. We need these GWAS, which in my fan understanding are basically brute force methods, to get enough statistical power to figure out what is going on.
This is vitally important to the future of the species. For example, our genomes have been building up genetic load over the past few generations. Genetic load is the accumulation of many rare mutations, almost all of which are bad. If we don't prune them ourselves, nature will eventually do it with a cataclysm.
Another example is the one that gets pushed in the news. We finally should be able to start entering the era of personalized medicine. The Human Genome Project hoped that it would come much sooner than it will end up coming, but human genetics turned out to be a lot more complex than they had hoped. It turns out that there seldom is a gene for something. In most cases, every something has many genes, and every gene contributes to multiple somethings.
I'm not sure I follow the hype. It sounds like we're essentially talking about machine learning, applied to medicine - which is huge and exciting, certainly. But is genome data a particularly useful signal? I mean, if you want an AI to detect cancer you probably want to give it x-rays, not genomes. Are there lots of counterexamples to that?
@Bill said:
This is vitally important to the future of the species. For example, our genomes have been building up genetic load over the past few generations. Genetic load is the accumulation of many rare mutations, almost all of which are bad. If we don't prune them ourselves, nature will eventually do it with a cataclysm.
I come to this conversation late, but it sounds like Bill is on the road to eugenics...
Bill has always been on the road to eugenics. Eugenics is a marvelous idea. Just because a bunch of madmen implemented it incorrectly doesn't mean it isn't essential to the future of the species.
You cannot repeal natural selection. We have the choice of maintaining the human collective genome in a kindly way, or nature will maintain it in a harsh way. There is no third option.
@fenomas said:
I'm not sure I follow the hype. It sounds like we're essentially talking about machine learning, applied to medicine - which is huge and exciting, certainly. But is genome data a particularly useful signal? I mean, if you want an AI to detect cancer you probably want to give it x-rays, not genomes. Are there lots of counterexamples to that?
The GWAS with lots of subjects will allow researchers to find "bad" genes. To the extent that disease is caused by genetics, this will be useful. Of course, a lot of disease is caused by hostile organisms.
Personally, I'm not as excited about the medical applications as I am about taking care of our accumulating genetic load. The genetic load issue, as we have just seen, makes people think of eugenics, which makes them all uncomfortable. Nevertheless, reducing the genetic load is essential.
My guess is that we will first start doing it through embryo selection. Next, we will start editing the actual germ cells. If I manage to live a normal lifespan, which is questionable, given my family history, I expect to see both before I die. Biotech doesn't make the headlines as often as electronics, but it is moving very fast, faster than the electronics revolution that we've all lived through.
@Bill said:
You cannot repeal natural selection. We have the choice of maintaining the human collective genome in a kindly way, or nature will maintain it in a harsh way. There is no third option.
Millenarianism correlates strongly with insanity. A rational person should eschew it outside spheres where they have extreme expertise.
Same answer on eugenics. "It only failed last time because madmen were in charge" is as compelling an argument against it as one can make.
What the article doesn't mention is that selecting embryos with the smallest number of rare mutations is equivalent to selecting the embryos that will grow up to be healthiest. I have faith that future parents will do the right thing.
Comments
The company above has now launched.
Also, for two days and the first one thousand orders, Veritas -- a separate company from the above -- is offering a 30X full genome sequencing for $199. Even with the costs of biotech plummeting, that is a great deal.
The name of the second company just gives me the heebie jeebies.
Probably just years of making Backup Exec work.
Finally! Genetics on the blockchain! Start the party!
In my estimation, this is a big deal. If the human race can round up enough volunteers to both get their genomes sequenced and allow researchers to study those sequences, they can start to do large (million genome plus) genome-wide association studies (GWAS).
We'll be able to start really getting things locked down. It turns out that a lot of the things that we are interested in aren't determined by a few genes of large effect, but by a large number of genes of small effect. We need these GWAS, which in my fan understanding are basically brute force methods, to get enough statistical power to figure out what is going on.
This is vitally important to the future of the species. For example, our genomes have been building up genetic load over the past few generations. Genetic load is the accumulation of many rare mutations, almost all of which are bad. If we don't prune them ourselves, nature will eventually do it with a cataclysm.
Another example is the one that gets pushed in the news. We finally should be able to start entering the era of personalized medicine. The Human Genome Project hoped that it would come much sooner than it will end up coming, but human genetics turned out to be a lot more complex than they had hoped. It turns out that there seldom is a gene for something. In most cases, every something has many genes, and every gene contributes to multiple somethings.
I'm not sure I follow the hype. It sounds like we're essentially talking about machine learning, applied to medicine - which is huge and exciting, certainly. But is genome data a particularly useful signal? I mean, if you want an AI to detect cancer you probably want to give it x-rays, not genomes. Are there lots of counterexamples to that?
I come to this conversation late, but it sounds like Bill is on the road to eugenics...
Bill has always been on the road to eugenics. Eugenics is a marvelous idea. Just because a bunch of madmen implemented it incorrectly doesn't mean it isn't essential to the future of the species.
You cannot repeal natural selection. We have the choice of maintaining the human collective genome in a kindly way, or nature will maintain it in a harsh way. There is no third option.
The GWAS with lots of subjects will allow researchers to find "bad" genes. To the extent that disease is caused by genetics, this will be useful. Of course, a lot of disease is caused by hostile organisms.
Personally, I'm not as excited about the medical applications as I am about taking care of our accumulating genetic load. The genetic load issue, as we have just seen, makes people think of eugenics, which makes them all uncomfortable. Nevertheless, reducing the genetic load is essential.
My guess is that we will first start doing it through embryo selection. Next, we will start editing the actual germ cells. If I manage to live a normal lifespan, which is questionable, given my family history, I expect to see both before I die. Biotech doesn't make the headlines as often as electronics, but it is moving very fast, faster than the electronics revolution that we've all lived through.
Millenarianism correlates strongly with insanity. A rational person should eschew it outside spheres where they have extreme expertise.
Same answer on eugenics. "It only failed last time because madmen were in charge" is as compelling an argument against it as one can make.
Bills vision, happening today: http://fortune.com/2018/11/26/chinese-researchers-genetic-engineering-babies-crispr-hiv/
Despite CRISPR baby controversy, Harvard University will begin gene-editing sperm
Year-old article: Eugenics 2.0: We’re at the Dawn of Choosing Embryos by Health, Height, and More
What the article doesn't mention is that selecting embryos with the smallest number of rare mutations is equivalent to selecting the embryos that will grow up to be healthiest. I have faith that future parents will do the right thing.