4 ms·

I guess I don't understand how OpenTrons will make a big difference. The pipetting robots we have around my company are the "expensive" ones by Hamilton and ot

I guess I don't understand how OpenTrons will make a big difference. The pipetting robots we have around my company are the "expensive" ones by Hamilton and other traditional makers. I think those machines cost $40k, which is definitely more expensive than OpenTrons, but they're hardly a limiting cost compared to generating samples and the disposables for whatever assay you're automating. Also, we have a department whose job is to run high-throughput assays, and they spend a lot of time and money validating an assay before they run the real samples through.

dekhn 7y ago
My feeling is that OpenTrons and similar systems would be a cost-effective way to scale up to warehouse-scale high throughput biology with more control over the process automation, making it easier to (for example) upload all the generated data to a large cloud and do large-scale machine learning and analysis over it. The model I like is basically the same as how Google scanned so many books so fast and cheaply- rather than buying the Hamilton equivalent, they built their own crappy scanners and trained the techs to deal with the crap. This was much more cost effective and scalable. but you need a ton of inhouse experts to deal with the crap hardware, and most large compannies don't want that- they just want a service that works.
- iso1337 7y ago
  I’ll give your analogy a try. Google could do that because the basic components were so cheap due to consumer imaging. On the lab side, there’s no similar downward pressure for making more accurate automated pipettes cheaper - I fully believe it’s possible, but the market isn’t there. Also, with imaging books, you can get away with a lot since ML/cheap labor/etc can be thrown at the distortion correction problem. With lab tests, you can’t really accept that same amount of error (ok, you can iff you understand the distortion and can process more samples to compensate). I do agree that most companies just want a service that works. The issue is that there’s no standardization. Depending on the assay you do, you will need wildly different setups and sensor systems. When you optimize whatever you are doing, often you have to re-tune your entire assay to bring it line with the new dynamic range of your output. If that weren’t the case, companies like Emerald and Transcriptic would have succeeded years ago. Selling things like DNA or cell lines have been the closest to a standard, scalable product. Otherwise, it’s just replacing armies of people at a CRO (Contract Research Org) with a custom project for automation and assaying (Zymergen’s model).
  - dekhn 7y ago
    At least in the area I work in (ML on fluorescence microscopy images used to estimate cellular and molecular phenotypes), the problem is extremely similar to the book reader that Google made. Heck, when I worked at Google I even built (20% time, open source) a prototype that could easily have been improved into what I describe (https://github.com/google/cncmicroscope-cad https://github.com/google/cncmicroscope-cad). We have to do the same kinds of distortion corrections (well, actually ours are far harder to do properly) to make the images useful for ML. I've always been sad that Emerald and Transcriptic haven't been insanely successful. I've tried to put my scopes into both systems (I know the founders), but never made any headway. I like OpenTrons better, but having spent enough of my time knee deep in tubing and stepper motors, I have more appreciation for well-engineered systems.
2YwaZHXV 7y ago
From my experience buying/getting quotes for Hamilton/Tecan machines, they're more likely in the $150-250k range. At least for the ones much wider and capable than that OpenTrons appears to be.
entee 7y ago
In grad school we bought robots because we thought they would make us faster. With very few exceptions they didn't, I was still faster with a multipipettor than our robots. This was because: 1.) Getting the robot to behave means it has to see the same thing every time, that's not easy to accomplish, hence the validation people do. 2.) Even when the thing is the same as before, robots break down. Handling solvents with robots is tricky because different solvents have different viscosities. If the robot misses a well, how do you know? If you have a team that knows the robot well, and it's set up well, then all these problems are manageable. But lets not kid ourselves that it's just trivially solved now because we have robots. An analogy in server land, we now have AWS and GCP. Basically now it's trivial to automate things it would take weeks/months to do by hand in the past. However, they each have an army of sysadmins and engineers behind the scenes servicing all the "robots" and making sure everything runs smoothly.
- dnautics 7y ago
  It's also possible your robot will be the robot equivalent of azure.
- ericmcer 7y ago
  So we should have just thrown away the first computers because inputting problems into them wasn't that much faster than doing it by hand?
  - mattkrause 7y ago
    Nobody's seriously suggesting that. That said, I do worry about all the hype and overpromising and whether that's eventually going to come back to haunt the entire field, like the (1st) AI Winter.
  - entee 7y ago
    Robots are absolutely useful, and major companies are building out massive robotics facilities to accelerate drug development. The point is that it's not a magic solution. It still takes a bunch of staff to run, and (much less appreciated) it takes a different mindset to be maximally useful. For example, maybe instead of doing a dozen informative but slightly complex assays, you do 1000 less useful but less complex assays. The end result might be just as useful if not more, but you have to structure your experiment differently. It's unlikely that technology will play out the same way in biotech as it has in pure tech. I'm heavily invested in the idea that it will be immensely useful, but given the different constraints and problem space I think the trajectory will be different.
dnautics 7y ago
Keep in mind: 40k is a limiting factor relative to the tiny amount of capital yc is willing to risk on any given team's hard science startup.
knyquist 7y ago
I do think that inexpensive liquid handling robots will make a big difference if they are used correctly. The expensive Hamilton robots are closer to ~$100k-$150k and require a ~$20k-$30k service contract per year. The opentrons robots are $4k total. That kind of price difference is a PhD-level scientist's salary. But you don't want to buy a cheap robot and soak up someone's time troubleshooting it. To effectively use something like an opentrons robot, you need a still-rare breed of scientist, someone who's both relatively competent at programming and also talented at the bench. This kind of person will become more and more common, but they are currently hard to find. Meanwhile, opentrons is working hard to lower the programming barrier-to-entry, but they're a ways off still.

dekhn 7y ago

My feeling is that OpenTrons and similar systems would be a cost-effective way to scale up to warehouse-scale high throughput biology with more control over the process automation, making it easier to (for example) upload all the generated data to a large cloud and do large-scale machine learning and analysis over it. The model I like is basically the same as how Google scanned so many books so fast and cheaply- rather than buying the Hamilton equivalent, they built their own crappy scanners and trained the techs to deal with the crap. This was much more cost effective and scalable. but you need a ton of inhouse experts to deal with the crap hardware, and most large compannies don't want that- they just want a service that works.

iso1337 7y ago
I’ll give your analogy a try. Google could do that because the basic components were so cheap due to consumer imaging. On the lab side, there’s no similar downward pressure for making more accurate automated pipettes cheaper - I fully believe it’s possible, but the market isn’t there. Also, with imaging books, you can get away with a lot since ML/cheap labor/etc can be thrown at the distortion correction problem. With lab tests, you can’t really accept that same amount of error (ok, you can iff you understand the distortion and can process more samples to compensate). I do agree that most companies just want a service that works. The issue is that there’s no standardization. Depending on the assay you do, you will need wildly different setups and sensor systems. When you optimize whatever you are doing, often you have to re-tune your entire assay to bring it line with the new dynamic range of your output. If that weren’t the case, companies like Emerald and Transcriptic would have succeeded years ago. Selling things like DNA or cell lines have been the closest to a standard, scalable product. Otherwise, it’s just replacing armies of people at a CRO (Contract Research Org) with a custom project for automation and assaying (Zymergen’s model).
- dekhn 7y ago
  At least in the area I work in (ML on fluorescence microscopy images used to estimate cellular and molecular phenotypes), the problem is extremely similar to the book reader that Google made. Heck, when I worked at Google I even built (20% time, open source) a prototype that could easily have been improved into what I describe (https://github.com/google/cncmicroscope-cad https://github.com/google/cncmicroscope-cad). We have to do the same kinds of distortion corrections (well, actually ours are far harder to do properly) to make the images useful for ML. I've always been sad that Emerald and Transcriptic haven't been insanely successful. I've tried to put my scopes into both systems (I know the founders), but never made any headway. I like OpenTrons better, but having spent enough of my time knee deep in tubing and stepper motors, I have more appreciation for well-engineered systems.

2YwaZHXV 7y ago

From my experience buying/getting quotes for Hamilton/Tecan machines, they're more likely in the $150-250k range. At least for the ones much wider and capable than that OpenTrons appears to be.

entee 7y ago

In grad school we bought robots because we thought they would make us faster. With very few exceptions they didn't, I was still faster with a multipipettor than our robots. This was because: 1.) Getting the robot to behave means it has to see the same thing every time, that's not easy to accomplish, hence the validation people do. 2.) Even when the thing is the same as before, robots break down. Handling solvents with robots is tricky because different solvents have different viscosities. If the robot misses a well, how do you know? If you have a team that knows the robot well, and it's set up well, then all these problems are manageable. But lets not kid ourselves that it's just trivially solved now because we have robots. An analogy in server land, we now have AWS and GCP. Basically now it's trivial to automate things it would take weeks/months to do by hand in the past. However, they each have an army of sysadmins and engineers behind the scenes servicing all the "robots" and making sure everything runs smoothly.

dnautics 7y ago
It's also possible your robot will be the robot equivalent of azure.
ericmcer 7y ago
So we should have just thrown away the first computers because inputting problems into them wasn't that much faster than doing it by hand?
- mattkrause 7y ago
  Nobody's seriously suggesting that. That said, I do worry about all the hype and overpromising and whether that's eventually going to come back to haunt the entire field, like the (1st) AI Winter.
- entee 7y ago
  Robots are absolutely useful, and major companies are building out massive robotics facilities to accelerate drug development. The point is that it's not a magic solution. It still takes a bunch of staff to run, and (much less appreciated) it takes a different mindset to be maximally useful. For example, maybe instead of doing a dozen informative but slightly complex assays, you do 1000 less useful but less complex assays. The end result might be just as useful if not more, but you have to structure your experiment differently. It's unlikely that technology will play out the same way in biotech as it has in pure tech. I'm heavily invested in the idea that it will be immensely useful, but given the different constraints and problem space I think the trajectory will be different.

dnautics 7y ago

Keep in mind: 40k is a limiting factor relative to the tiny amount of capital yc is willing to risk on any given team's hard science startup.

knyquist 7y ago

I do think that inexpensive liquid handling robots will make a big difference if they are used correctly. The expensive Hamilton robots are closer to ~$100k-$150k and require a ~$20k-$30k service contract per year. The opentrons robots are $4k total. That kind of price difference is a PhD-level scientist's salary. But you don't want to buy a cheap robot and soak up someone's time troubleshooting it. To effectively use something like an opentrons robot, you need a still-rare breed of scientist, someone who's both relatively competent at programming and also talented at the bench. This kind of person will become more and more common, but they are currently hard to find. Meanwhile, opentrons is working hard to lower the programming barrier-to-entry, but they're a ways off still.