5 ms·

There's an interview with Ken Thompson from about 2008 where he says that most of the work in CS has already been done, and he's advising his son to go into bio

There's an interview with Ken Thompson from about 2008 where he says that most of the work in CS has already been done, and he's advising his son to go into biology. I feel like CRISPR is the transistor of the 21st century.

jazzyjackson 4y ago
I'm still holding out for photonics and other optical-analog computers (where are my instantaneous trig co-processors?) but that does sound like good advice
Invictus0 4y ago
AI?
- JKCalhoun 4y ago
  Perhaps also future biology?
shimon 4y ago
I studied CS and now work in software systems for biomedical research. It's difficult to overstate how different the fields are, so I don't entirely agree with this statement. But I do agree there are going to be lots and lots of huge discoveries in biology in the 21st century. The main difference is that CS attempts to generate and study complex systems built from well-understood components, whereas biology attempts to understand and manipulate systems that evolved naturally over eons. Imagine dropping a fully functional internet-connected Google Home Hub into 1960-era humanity and asking them to figure out how it works so they modify it to sound like Walter Cronkite. There are thousands of problems on this order of complexity in biology. It's wild.
- tyre 4y ago
  think of the tech debt in our legacy codebase
  - elevaet 4y ago
    3.7 billion years of refactoring has kept it pretty clean and functional. We'll need to do a shit ton of unit and integration testing before we commit changes.
  - redtexture 4y ago
    There is a lot of pruning that occurs, evolutionarily speaking, and a lot of what was thought to be "useless" genome has been discovered to be conserved over generations, and that there is use for that part of the genome.
- mannymanman 4y ago
  What was your path going from CS -> bio? Interested in a similar path
  - pddpro 4y ago
    Same here. The more I've progressed in CS, the more dissatisfied I am. Outside of creating algorithms that vie for constant user attention (the basic business model of FAANG), I don't see any fruitful application of my skills. I'd much rather move towards the domains where my knowledge of data, systems, and algorithms could be better utilized (medicine, Genomics, structural engineering, governance etc).
    - Balgair 4y ago
      I've done the physics->neuro leap, so I may be of some use here. The path is pretty clear, but takes time. Essentially, you need to go back to school and learn biology. Fortunately, many grad programs in the US are desperate for people that want to be trained as biologists but have relevant skills in other areas like CS. So skip going back to undergrad and just apply to grad programs. Unfortunately, that means you have to join the Ivory Tower's horrible system for a while. A 'good' tactic is to get into a PhD program where you'll be paid, learn everything, get your MS, and then quit the program after ~3 years with a free MS. Fair warning, the learning will be absolutely horrible and you'll be on the bubble of being kicked out; it really is that much info you're trying to digest in such a short time period. But if you're not worried about scholarships and grades, then that's fine. Your PI will hate you, but then again they hate everyone, so it's a wash. If you're serious about grad school then read this first: https://acoup.blog/2021/10/01/collections-so-you-want-to-go-to-grad-school-in-the-academic-humanities/ https://acoup.blog/2021/10/01/collections-so-you-want-to-go-... One thing to be clear about though, jobs in biotech are much less well paid than in CS. You're looking at a 1/3rd to 1/4th salary decrease for pure bio jobs as compared to programmer jobs. Even leveraging your coding skills for biotech companies is going to be tough; you'll be pigeon holed into either a lab role or a coder role. The true blended roles are very rare. So much so as that you'll essentially have to start your own company, or be the heart of any company your join. So, good money there, but huge pressures.
- pddpro 4y ago
  How difficult is it to transition from Computer Science to Biomedics? Particularly towards the field of Genomics (where CRISPR is).
  - Balgair 4y ago
    Having done the physics -> neuro leap, it's pretty tough. You have to learn a whole new set of fields and new ways of thinking. That takes time. To be 'good' at genomics, you kinda need to know how the genes are implemented in the various model organisms. Which means you need to know the relevant biology, biochemistry, chemistry, and physics of the situations. That's, essentially, an entire undergrad education. Then, you get to do the actual work, which takes about 1.5 years of study, so most of a masters degree. Then you can start really doing the work. For me, the first big realization coming from physics was that these little yeast cells and zebrafish aren't just little machines of quantum chemistry. They really are alive, even down to the cellular level, and they are studying you too. There were hundreds of such insights.
foobiekr 4y ago
The problem with this is that biology will likely end up dominated by China due to a willingness to conduct experiments that are otherwise non-viable in most countries.
- ornornor 4y ago
  > non-viable Did you mean unethical or are you talking about something else?
  - mod 4y ago
    Forbidden by law or fear of backlash (often due to ethical implications)
  - foobiekr 4y ago
    I mean not permitted by US and European standards - legal, social and ethical.
- jeffreyrogers 4y ago
  There's still so much basic research to be done that I doubt this will be a limiting factor for a while.
jarenmf 4y ago
Yeah, it feels like most fields are stagnated except for biology and neuroscience. I am a postdoc right now but have considered seriously switching fields to work on something exciting.
- PartiallyTyped 4y ago
  I fail to see how CS has stagnated since 2008, so I hope somebody could illuminate it for me.
dekhn 4y ago
CRISPR is more like a machine to make large-node-size integrated circuits. Somebody else has to design the circuit, make it manufacturable, and integrate the circuit with a whole bunch of other hardware. I work in biotech at a company that is one of the few golden geese that lays 2-3 successful drugs with no competitors every few years. I have 30+ years of experience (deep experience) in machine learning, biology, and computer science. We are so far behind where we could be, in terms of turning biology into technology, that's almost shameful. Every day I see another system that says it can generate 10 times the data of the previous machine, but the actual amount of knowledge we are extracting for all that data collection is growing logarithmically. This is because for a long time biology has greatly underfunded computing and data. The one great shining light is AlphaFold. AF2 finally demonstrated to a wide range of scientists across many domains that a really great team using techniques that are barely known outside of FAAMG can work with some long-term experts to move a metric (quality of predicted protein structures compared to golden data) substantially further and faster than even the most wildly optimistic predicted. Not only that, some of the techniques they used didn't even exist several years ago (transformers, jax, various graph learning systems), and the work was replicated externally once the leading academic team had a hint of the direction to go in. To me, nothing about what I said is surprising to me; I predicted these outcomes a long time ago. Most of the reasons that it comes slower than it could are combinations of culture, incentive, morals/ethics, politics, innovator's dilemmas and a hundred different bottlenecks. Recently, the challenge has been that most of the really smart computational biologists disappear into FAAMG and don't contribute back the things they learn there to research. We're all waiting for that next moment when the cross product of Genentech and Isomorphic Labs announces that they have a computational model that can do end to end prediction of drug, from initial disease target to FDA approval post-phase III trial. That's been the dream for some time but we're nowhere near it still, and it remains to be seen whether some group can conjure all the necessary bits to solve the remaining underlying problems associated with that "far beyond NP-hard problem"
- printf42 4y ago
  Shouldn't it be MAANG now?
  - 4y ago
seydor 4y ago
Biology is being automated by startups that are doing the experiments for you, AWS-style. Looks like software is going to be eating biology as well
- mannymanman 4y ago
  any examples?
  - kyawzazaw 4y ago
    Radix.bio

jazzyjackson 4y ago

I'm still holding out for photonics and other optical-analog computers (where are my instantaneous trig co-processors?) but that does sound like good advice

Invictus0 4y ago

AI?

JKCalhoun 4y ago
Perhaps also future biology?

shimon 4y ago

I studied CS and now work in software systems for biomedical research. It's difficult to overstate how different the fields are, so I don't entirely agree with this statement. But I do agree there are going to be lots and lots of huge discoveries in biology in the 21st century. The main difference is that CS attempts to generate and study complex systems built from well-understood components, whereas biology attempts to understand and manipulate systems that evolved naturally over eons. Imagine dropping a fully functional internet-connected Google Home Hub into 1960-era humanity and asking them to figure out how it works so they modify it to sound like Walter Cronkite. There are thousands of problems on this order of complexity in biology. It's wild.

tyre 4y ago
think of the tech debt in our legacy codebase
- elevaet 4y ago
  3.7 billion years of refactoring has kept it pretty clean and functional. We'll need to do a shit ton of unit and integration testing before we commit changes.
- redtexture 4y ago
  There is a lot of pruning that occurs, evolutionarily speaking, and a lot of what was thought to be "useless" genome has been discovered to be conserved over generations, and that there is use for that part of the genome.
mannymanman 4y ago
What was your path going from CS -> bio? Interested in a similar path
- pddpro 4y ago
  Same here. The more I've progressed in CS, the more dissatisfied I am. Outside of creating algorithms that vie for constant user attention (the basic business model of FAANG), I don't see any fruitful application of my skills. I'd much rather move towards the domains where my knowledge of data, systems, and algorithms could be better utilized (medicine, Genomics, structural engineering, governance etc).
  - Balgair 4y ago
    I've done the physics->neuro leap, so I may be of some use here. The path is pretty clear, but takes time. Essentially, you need to go back to school and learn biology. Fortunately, many grad programs in the US are desperate for people that want to be trained as biologists but have relevant skills in other areas like CS. So skip going back to undergrad and just apply to grad programs. Unfortunately, that means you have to join the Ivory Tower's horrible system for a while. A 'good' tactic is to get into a PhD program where you'll be paid, learn everything, get your MS, and then quit the program after ~3 years with a free MS. Fair warning, the learning will be absolutely horrible and you'll be on the bubble of being kicked out; it really is that much info you're trying to digest in such a short time period. But if you're not worried about scholarships and grades, then that's fine. Your PI will hate you, but then again they hate everyone, so it's a wash. If you're serious about grad school then read this first: https://acoup.blog/2021/10/01/collections-so-you-want-to-go-to-grad-school-in-the-academic-humanities/ https://acoup.blog/2021/10/01/collections-so-you-want-to-go-... One thing to be clear about though, jobs in biotech are much less well paid than in CS. You're looking at a 1/3rd to 1/4th salary decrease for pure bio jobs as compared to programmer jobs. Even leveraging your coding skills for biotech companies is going to be tough; you'll be pigeon holed into either a lab role or a coder role. The true blended roles are very rare. So much so as that you'll essentially have to start your own company, or be the heart of any company your join. So, good money there, but huge pressures.
pddpro 4y ago
How difficult is it to transition from Computer Science to Biomedics? Particularly towards the field of Genomics (where CRISPR is).
- Balgair 4y ago
  Having done the physics -> neuro leap, it's pretty tough. You have to learn a whole new set of fields and new ways of thinking. That takes time. To be 'good' at genomics, you kinda need to know how the genes are implemented in the various model organisms. Which means you need to know the relevant biology, biochemistry, chemistry, and physics of the situations. That's, essentially, an entire undergrad education. Then, you get to do the actual work, which takes about 1.5 years of study, so most of a masters degree. Then you can start really doing the work. For me, the first big realization coming from physics was that these little yeast cells and zebrafish aren't just little machines of quantum chemistry. They really are alive, even down to the cellular level, and they are studying you too. There were hundreds of such insights.

foobiekr 4y ago

The problem with this is that biology will likely end up dominated by China due to a willingness to conduct experiments that are otherwise non-viable in most countries.

ornornor 4y ago
> non-viable Did you mean unethical or are you talking about something else?
- mod 4y ago
  Forbidden by law or fear of backlash (often due to ethical implications)
- foobiekr 4y ago
  I mean not permitted by US and European standards - legal, social and ethical.
jeffreyrogers 4y ago
There's still so much basic research to be done that I doubt this will be a limiting factor for a while.

jarenmf 4y ago

Yeah, it feels like most fields are stagnated except for biology and neuroscience. I am a postdoc right now but have considered seriously switching fields to work on something exciting.

PartiallyTyped 4y ago
I fail to see how CS has stagnated since 2008, so I hope somebody could illuminate it for me.

dekhn 4y ago

CRISPR is more like a machine to make large-node-size integrated circuits. Somebody else has to design the circuit, make it manufacturable, and integrate the circuit with a whole bunch of other hardware. I work in biotech at a company that is one of the few golden geese that lays 2-3 successful drugs with no competitors every few years. I have 30+ years of experience (deep experience) in machine learning, biology, and computer science. We are so far behind where we could be, in terms of turning biology into technology, that's almost shameful. Every day I see another system that says it can generate 10 times the data of the previous machine, but the actual amount of knowledge we are extracting for all that data collection is growing logarithmically. This is because for a long time biology has greatly underfunded computing and data. The one great shining light is AlphaFold. AF2 finally demonstrated to a wide range of scientists across many domains that a really great team using techniques that are barely known outside of FAAMG can work with some long-term experts to move a metric (quality of predicted protein structures compared to golden data) substantially further and faster than even the most wildly optimistic predicted. Not only that, some of the techniques they used didn't even exist several years ago (transformers, jax, various graph learning systems), and the work was replicated externally once the leading academic team had a hint of the direction to go in. To me, nothing about what I said is surprising to me; I predicted these outcomes a long time ago. Most of the reasons that it comes slower than it could are combinations of culture, incentive, morals/ethics, politics, innovator's dilemmas and a hundred different bottlenecks. Recently, the challenge has been that most of the really smart computational biologists disappear into FAAMG and don't contribute back the things they learn there to research. We're all waiting for that next moment when the cross product of Genentech and Isomorphic Labs announces that they have a computational model that can do end to end prediction of drug, from initial disease target to FDA approval post-phase III trial. That's been the dream for some time but we're nowhere near it still, and it remains to be seen whether some group can conjure all the necessary bits to solve the remaining underlying problems associated with that "far beyond NP-hard problem"

printf42 4y ago
Shouldn't it be MAANG now?
- 4y ago

seydor 4y ago

Biology is being automated by startups that are doing the experiments for you, AWS-style. Looks like software is going to be eating biology as well

mannymanman 4y ago
any examples?
- kyawzazaw 4y ago
  Radix.bio