will that disassembly tool include regulator genes and all the protein-protein interactions that influence said regulator genes, ad infinitum? there's so much more complexity to life than gene sequences themselves.
Virus are like 30k base pairs. It's a quite short sequence for neural networks to tackle.
You can already use GPT-2 to generate compiled code already. You train it on known viruses and have it predict new variants.
You can also for better sample efficiency, jointly train to predict for any measurable characteristics you like, for example mortality or propagation speed, you then synthesize the various proposed DNA sequences, and inject them into hosts. Compare the results with the prediction and update your models accordingly.
Of course you will have plenty of non functioning organism, but that's probably a more efficient way of exploring this search space than what nature does with random mutations and cross-overs.
It's like giving virus the ability to virtually meet-up in the hall of fame to evolve faster.
I can see its power even if I'm not a biologist. For example, I just submitted (everyone can use the system) a BLAST task for a megablast search to compare NC_045512.2 (Wuhan-Hu-1) and NC_004718.3 (SARS coronavirus) for an alignment, and the system is able to "diff" between the two sequences (and even has an option to visualize it in a 2D graph), and finally telling us that 82.3% of the both genetic sequences are identical (not really close in a biological sense). You can also search identical genetic sequences across the entire databases by constructing a more complex query.
And apparently, the specific algorithm used to produce my result is described in this paper.
> Zheng Zhang, Scott Schwartz, Lukas Wagner, and Webb Miller (2000), "A greedy algorithm for aligning DNA sequences", J Comput Biol 2000; 7(1-2):203-14.
It looks like a simple algorithm, which can be implemented for fun as a weekend project. One can run one's program on real DNA sequences, and compare the outputs with NIH's implementation. Don't forget to write an article and submit it to Hacker News when you're done. Biohacking!
There are DNA based virus detection techniques that take fragments of DNA of unknown origin and compare them to a database of known bad stuff. So I think it could allow people that have the ability to sequence DNA to identify this virus if they have, for example, a patient.
What's interesting is Wuhan is home to a virology research center. They were literally testing bats inside of Wuhan for viral diseases. This makes me think this strain broke out of the lab. From the above article:
"...to clinch the case, a team led by Shi Zheng-Li and Cui Jie of the Wuhan Institute of Virology in China sampled thousands of horseshoe bats in locations across the country"
i am not for sure that's more likely than the situation of people eating infected bats or other animals infected by bats. the chinese do not have great hygiene when it comes to food sourcing, preparation, and consumption. the chinese have all sorts of "beliefs" about food and are very stubborn to accept change. for example, thinking hot water as a sort of elixir and cleanser (e.g., killing parasites). oh look, it's served in hot water, so it's fine. also, the shared eating style (i.e., using a lazy susan) is not exactly a great way to contain illness, even if the food itself is fine.
Well, to be fair to them, in America, we have rural hicks in the midwest hunting rabbits, squirrels, snakes, hamsters, and other mammals for recreational food. These are wild animals that possibly carry rabies and other diseases. And didn’t the Hantavirus from the 90s originate from infected rodents in the United States?
Do you want to educate these fellow Americans to not consume such possibly infected wildlife? Because surely, they live in America, and they ought to know better.
And if you’re on a deserted island, then you’d probably eat these animals too, if you wanted to survive.
And sharing food on the lazy Susan is perfectly fine, just use a sterile serving spoon, and don’t lick the handle. Don’t you eat dinner this way with your own family meals?
Currently, the scientists don’t know yet where this vector originated from. And they are locking down the city out of an abundance of caution.
Perhaps this episode will force them to implement strict meat sourcing requirements. Such as, all meat sold for consumption, must come from a farm, and pass a food handling certification process.
This is clearly a very expensive operation and administration, but they just quarantined a city, so that is an even more expensive and politically damaging event.
The first step in preventing such an outbreak from happening again in the future, is in personal education, and a reinforcement of following basic hygiene practices. Perhaps this event will finally trigger the China central government to mandate such a national education process for everyone.
the rural hicks you mentioned are an extremely small minority in the u.s., found in the backwoods of louisiana and kentucky and the like. the majority of people find this practice uncouth. the one food style i mentioned is overwhelmingly the majority in china.
> And sharing food on the lazy Susan is perfectly fine, just use a sterile serving spoon, and don’t lick the handle. Don’t you eat dinner this way with your own family meals?
i have never seen ANY chinese meal served from a lazy susan use individual serving spoons for the individual meals. everyone uses their own chopsticks to grab the food and put it in their plate or they simply eat directly from the shared plate. if there is a serving spoon for a dish, sometimes provided by restaurants, individual chopsticks are still used to help.
it is a rather major faux pas in american eating to use any utensil used by an individual to touch or serve a shared dish. it is overwhelmingly the case that a serving spoon is dedicated to a dish and is only used to move the food to an individual's plate.
now, i would imagine this culture difference of eating is probably a minor one in the grand scheme of food hygeine (preparation and sourcing being more important i think), but i think it's pretty easy to say one is healthier than another.
in china, it is a constant battle to ask and understand how a dish was sourced and prepared, as most of it is taken as is by chinese. the eating habits are unchangeable at a local level and one must deal with it.
my original comment was on the likelihood of this originating in a market versus a lab-based outbreak. of course most chinese do not eat bat. i didn't imply they did. the comment you replied to was me dismissing the commenter's argument. but the wuhan market and its patrons do eat bats and other animals that could already be infected by bats. so it's not far fetched that the virus did indeed originate there vs a lab leak.
the latter part of my comment was regarding the potential spread of disease. your comment addresses someone who already knows they are sick. the onset period of this virus is one to two weeks. that's a minimum of 14-21 meals before any symptoms show up, leaving plenty of time for person to person transmission. that plus the density of china's major cities and this particular time of year regarding travel yields a rather fertile ground for disease spread.
A recent Time article [1] is theorizing that the source may come from wet markets.
It also theorized that the source of the SARS virus might have originated from an infected cat. And Ebola may have originated from infected bat feces, that might have landed on a girl’s toy, which she then put into her mouth. And the MERS virus in the Middle East may have come from an infected camel.
This appears to be a logical and scientific deduction.
So for this Coronavirus, it appears to follow the same pattern. Infected animal feces mutated into some virus, that made the transmission to humans.
My theory, is that perhaps an infected animal’s feces, may have mixed with other fluids, like blood, from freshly slaughtered animals, and mutated into something that infected a human with a weakened immune system (easy target), who then transmitted it to other healthy humans by contact.
On one hand, it does seem that the best way to guarantee that you’re sourcing your meat from a fresh and healthy animal, is to pick the animal that you want, and to slaughter it on the spot. This means that the meat vendor does not have to maintain a refrigeration system, and you don’t introduce other chemicals or salts to preserve your meat.
But the problem is that you have to keep the animals in captivity, and the meat vendor must handle, and clean all their urine and feces. And when animals get sick, their feces and urine gets infected, thus you increase the probability of an outbreak like this.
What’s the solution? Maybe China must begin eliminating wet markets, and eliminate the practice of keeping live animals in captivity, in order to minimize the incubation, mutation, and the transmission of such viruses. And instead, to begin using a refrigeration system, to preserve slaughtered animals.
"Bat hunt
...the researchers spent five years monitoring the bats that lived there, collecting fresh guano and taking anal swabs[1]. "
It shows that the researchers are just taking some samples back to the lab. In the Chinese version of the report (http://www.xinhuanet.com/tech/2017-12/13/c_1122101425.htm) says "作为课题组长，石正丽时常带队爬山钻洞。采样工作通常是4人一组。组员戴着N95口罩、手套和头灯，穿着外套，傍晚时在蝙蝠洞口支起捕鸟网。夜里，他们从捕鸟网上取下落网的蝙蝠，在野外临时搭起的工作台上，连夜进行肛拭子取样。他们捉住蝙蝠，用浸过生理盐水的棉签拭子，插入蝙蝠肛门拭取样品。提取肛拭子之后，蝙蝠会被放走。“石老师一直坚持无侵害取样，尽量不剖杀蝙蝠，采样时也尽量减少对蝙蝠的身体伤害。”罗东升说。有些蝙蝠种群数量很少，破坏性的采样很有可能对整个种群造成毁灭性打击。蝙蝠在维护生态平衡中发挥重要作用，捕杀大量蝙蝠，将对生态系统造成重大损失。" translated by Google: "As the team leader, Shi Zhengli often led the team to climb the mountain to drill holes. Sampling is usually performed in groups of four. The team members wore N95 masks, gloves and headlights, and coats, and in the evening set up a bird-catching net at the bat hole. At night, they took the bats of the falling net from the birding nets, and performed anal swab sampling on the workbench temporarily set up in the wild overnight. They grabbed the bat and used a cotton swab swab soaked in normal saline to insert the bat's anus to swab the sample.
After the anal swab has been extracted, the bat will be released. "Mr. Shi has always insisted on non-invasive sampling, try not to kill the bats, and try to reduce the physical damage to the bats as much as possible," Luo Dongsheng said. Some bat populations are small, and destructive sampling is likely to have a devastating impact on the entire population. Bats play an important role in maintaining the ecological balance, killing a large number of bats will cause significant losses to the ecosystem."
Yet it is clear that a lot of so called SEA FOOD markets are semi-secretly selling wild animals.
21 comments
[ 4.5 ms ] story [ 69.8 ms ] threadYou can already use GPT-2 to generate compiled code already. You train it on known viruses and have it predict new variants.
You can also for better sample efficiency, jointly train to predict for any measurable characteristics you like, for example mortality or propagation speed, you then synthesize the various proposed DNA sequences, and inject them into hosts. Compare the results with the prediction and update your models accordingly.
Of course you will have plenty of non functioning organism, but that's probably a more efficient way of exploring this search space than what nature does with random mutations and cross-overs.
It's like giving virus the ability to virtually meet-up in the hall of fame to evolve faster.
* Run BLAST
* Pick Primers
* Highlight Sequence Features
* Find in this Sequence
The BLAST algorithm is an interesting read. https://en.wikipedia.org/wiki/BLAST_(biotechnology)
I can see its power even if I'm not a biologist. For example, I just submitted (everyone can use the system) a BLAST task for a megablast search to compare NC_045512.2 (Wuhan-Hu-1) and NC_004718.3 (SARS coronavirus) for an alignment, and the system is able to "diff" between the two sequences (and even has an option to visualize it in a 2D graph), and finally telling us that 82.3% of the both genetic sequences are identical (not really close in a biological sense). You can also search identical genetic sequences across the entire databases by constructing a more complex query.
Here's my result: https://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Get&RID=2MDW41R...
And apparently, the specific algorithm used to produce my result is described in this paper.
> Zheng Zhang, Scott Schwartz, Lukas Wagner, and Webb Miller (2000), "A greedy algorithm for aligning DNA sequences", J Comput Biol 2000; 7(1-2):203-14.
http://pipmaker.bx.psu.edu/dist/greedy.pdf
It looks like a simple algorithm, which can be implemented for fun as a weekend project. One can run one's program on real DNA sequences, and compare the outputs with NIH's implementation. Don't forget to write an article and submit it to Hacker News when you're done. Biohacking!
Bat cave solves mystery of deadly SARS virus — and suggests new outbreak could occur (2017): https://www.nature.com/articles/d41586-017-07766-9
"...to clinch the case, a team led by Shi Zheng-Li and Cui Jie of the Wuhan Institute of Virology in China sampled thousands of horseshoe bats in locations across the country"
Do you want to educate these fellow Americans to not consume such possibly infected wildlife? Because surely, they live in America, and they ought to know better.
And if you’re on a deserted island, then you’d probably eat these animals too, if you wanted to survive.
And sharing food on the lazy Susan is perfectly fine, just use a sterile serving spoon, and don’t lick the handle. Don’t you eat dinner this way with your own family meals?
Currently, the scientists don’t know yet where this vector originated from. And they are locking down the city out of an abundance of caution.
Perhaps this episode will force them to implement strict meat sourcing requirements. Such as, all meat sold for consumption, must come from a farm, and pass a food handling certification process.
This is clearly a very expensive operation and administration, but they just quarantined a city, so that is an even more expensive and politically damaging event.
The first step in preventing such an outbreak from happening again in the future, is in personal education, and a reinforcement of following basic hygiene practices. Perhaps this event will finally trigger the China central government to mandate such a national education process for everyone.
> And sharing food on the lazy Susan is perfectly fine, just use a sterile serving spoon, and don’t lick the handle. Don’t you eat dinner this way with your own family meals?
i have never seen ANY chinese meal served from a lazy susan use individual serving spoons for the individual meals. everyone uses their own chopsticks to grab the food and put it in their plate or they simply eat directly from the shared plate. if there is a serving spoon for a dish, sometimes provided by restaurants, individual chopsticks are still used to help.
it is a rather major faux pas in american eating to use any utensil used by an individual to touch or serve a shared dish. it is overwhelmingly the case that a serving spoon is dedicated to a dish and is only used to move the food to an individual's plate.
now, i would imagine this culture difference of eating is probably a minor one in the grand scheme of food hygeine (preparation and sourcing being more important i think), but i think it's pretty easy to say one is healthier than another.
in china, it is a constant battle to ask and understand how a dish was sourced and prepared, as most of it is taken as is by chinese. the eating habits are unchangeable at a local level and one must deal with it.
I don’t know a single Chinese person that has eaten bat before, and I’ve been going to the country every year for 20 years now.
It’s only because China has such a large population that even 1% is 10 million people.
If anyone is sick, I ALWAYS see them using a separate set of chopsticks to get food, and they usually will put it in a separate plate.
You eat with family and friends and a sick person would take extra precautions to not infect them.
the latter part of my comment was regarding the potential spread of disease. your comment addresses someone who already knows they are sick. the onset period of this virus is one to two weeks. that's a minimum of 14-21 meals before any symptoms show up, leaving plenty of time for person to person transmission. that plus the density of china's major cities and this particular time of year regarding travel yields a rather fertile ground for disease spread.
It also theorized that the source of the SARS virus might have originated from an infected cat. And Ebola may have originated from infected bat feces, that might have landed on a girl’s toy, which she then put into her mouth. And the MERS virus in the Middle East may have come from an infected camel.
This appears to be a logical and scientific deduction.
So for this Coronavirus, it appears to follow the same pattern. Infected animal feces mutated into some virus, that made the transmission to humans.
My theory, is that perhaps an infected animal’s feces, may have mixed with other fluids, like blood, from freshly slaughtered animals, and mutated into something that infected a human with a weakened immune system (easy target), who then transmitted it to other healthy humans by contact.
On one hand, it does seem that the best way to guarantee that you’re sourcing your meat from a fresh and healthy animal, is to pick the animal that you want, and to slaughter it on the spot. This means that the meat vendor does not have to maintain a refrigeration system, and you don’t introduce other chemicals or salts to preserve your meat.
But the problem is that you have to keep the animals in captivity, and the meat vendor must handle, and clean all their urine and feces. And when animals get sick, their feces and urine gets infected, thus you increase the probability of an outbreak like this.
What’s the solution? Maybe China must begin eliminating wet markets, and eliminate the practice of keeping live animals in captivity, in order to minimize the incubation, mutation, and the transmission of such viruses. And instead, to begin using a refrigeration system, to preserve slaughtered animals.
[1] https://time.com/5770904/wuhan-coronavirus-wild-animals/
It shows that the researchers are just taking some samples back to the lab. In the Chinese version of the report (http://www.xinhuanet.com/tech/2017-12/13/c_1122101425.htm) says "作为课题组长，石正丽时常带队爬山钻洞。采样工作通常是4人一组。组员戴着N95口罩、手套和头灯，穿着外套，傍晚时在蝙蝠洞口支起捕鸟网。夜里，他们从捕鸟网上取下落网的蝙蝠，在野外临时搭起的工作台上，连夜进行肛拭子取样。他们捉住蝙蝠，用浸过生理盐水的棉签拭子，插入蝙蝠肛门拭取样品。提取肛拭子之后，蝙蝠会被放走。“石老师一直坚持无侵害取样，尽量不剖杀蝙蝠，采样时也尽量减少对蝙蝠的身体伤害。”罗东升说。有些蝙蝠种群数量很少，破坏性的采样很有可能对整个种群造成毁灭性打击。蝙蝠在维护生态平衡中发挥重要作用，捕杀大量蝙蝠，将对生态系统造成重大损失。" translated by Google: "As the team leader, Shi Zhengli often led the team to climb the mountain to drill holes. Sampling is usually performed in groups of four. The team members wore N95 masks, gloves and headlights, and coats, and in the evening set up a bird-catching net at the bat hole. At night, they took the bats of the falling net from the birding nets, and performed anal swab sampling on the workbench temporarily set up in the wild overnight. They grabbed the bat and used a cotton swab swab soaked in normal saline to insert the bat's anus to swab the sample.
After the anal swab has been extracted, the bat will be released. "Mr. Shi has always insisted on non-invasive sampling, try not to kill the bats, and try to reduce the physical damage to the bats as much as possible," Luo Dongsheng said. Some bat populations are small, and destructive sampling is likely to have a devastating impact on the entire population. Bats play an important role in maintaining the ecological balance, killing a large number of bats will cause significant losses to the ecosystem."
Yet it is clear that a lot of so called SEA FOOD markets are semi-secretly selling wild animals.
https://news.ycombinator.com/item?id=22130806