The BioBricks Foundation, an organization bringing the spirit of open source to genetic programming, engaged me to develop Public Domain Chronicle, a fast, easy, and free way to secure scientific methods and findings for the public domain.
Public Domain Chronicle combines elements of defensive publication, open-access scholarship, and commons licensing to make public domain advocates the fastest runners in the race to publish, preventing others from patenting their findings.
PDC's disclosure form for findings in biology is shorter and easier than any standard corporate or academic invention-disclosure process, and produces an immediate, public, republishable prior art record.
It's early days for the project, and we're seeking out as many potential researcher-contributors as possible. We're particularly keen to hear from academic scientists and folks in corporate tech transfer offices who may prefer PDC to expensive defensive publication services.
Some critical questions: who is the target audience for these instruments? Aren't academic budgets sufficient to provide for a lab with proper basic instruments? Are these kind of instruments a financial bottleneck when setting up a lab? And do we want biology PhD students and postdocs to spend time on instrument-making (and publish papers on that) rather than on biology?
Great questions. The target audience for these instruments are scientists and hobbyists who don't want to spend a ton of money on similar commercial systems, and want the flexibility to modify their operation (e.g., poseidon can run custom flow profiles per experiment whereas off-the-shelf commercial systems typically only run one flow rate per experiment). The purpose of poseidon is to show that open source biological instruments can be developed and used by a community, similarly to how open source software tools in biology are developed and used.
Academic budgets vary from institution to institution and are sometimes determined by exogenous forces beyond the lab's control.
A complete commercial system to do single-cell RNA sequencing costs tens of thousands of dollars! Using alternatives such as the Harvard Apparatus syringe pumps and DropSeq [0] to run the same experiment will still cost you into the thousands of dollars. With the poseidon system, we greatly reduce these costs. Users can build the instruments to run these experiments for less than $400 and are not restricted to additional costs and tedious firmware upgrades to expand the system.
In response to your point on time management and instrument-making, I think that if there exists a need to develop these systems such that they will advance biological experiments then it's totally cool to have academics work on these sorts of projects! Biologists and bioengineers have always developed tools alongside discovery and this is no different from developing bioinformatics tools.
Thanks for these great answers. The only point where I would disagree is your comparison with bioinformatics software. The difference there is that installing a bioinformatics package and learning to use it takes far less time than crafting an instrument, especially considering tasks like finding the right suppliers of components, and calibrating and testing the device (for which you might need other instruments). But on the other hand, perhaps building these self-made instruments can become a task for specialized, facilitating instrument-makers that are already part of many institutions.
The poseidon system was explicitly designed with ease of assembly in mind. If you look at the build videos [1] you'll see that assembly of the entire system (3 pumps + microscope station) takes less than an hour and requires just pliers and screwdrivers.
The importance of ease of assembly was a lesson that we took from assembling the miniDrops microfluidics station [2] developed specifically for one kind of experiment (dropSeq) [3]. The miniDrops is very good at what it does but assembly was somewhat cumbersome: it required ordering a custom PCB, specialized parts only available from one vendor (whom I had to nag over the phone to send me a quote!) and assembly of the device itself took 10-20h.
Not every kind of equipment can be made as easy to source and assemble as we did with poseidon, but we really think that keeping this at the front of your mind can make or break the adoption of a piece of open source hardware. This is especially true in the context of biology laboratories, where many people are not what you could call "hackers" or "makers" and will be immediately put off by a daunting assembly process.
> And do we want biology PhD students and postdocs to spend time on instrument-making (and publish papers on that) rather than on biology?
This is an age old question. And the answer is: yes, we actually do.
Even if a lab can afford all its instruments, sometimes you are researching something that requires a different approach. Most commercially available lab machines are NOT hackable, and you will have a difficult time bending them into a different configuration.
Yeah, for my grad thesis I had to figure out how to incorporate a fuel injector into a chemical reactor to meter fluid flow. Novel research equipment is by definition not off-the-shelf, so the skills to roll your own is often necessary to push the envelope.
And a lot of the time that equipment isn't funded by a current project and you need to prove the idea is sound before you can justify asking for the money to do it. Of course if that proof of concept is good enough to keep being used all the better.
We couldn't find a programmable temperature controller for a microscope that met our requirements so I ended up building it from scratch. It took me a month and I didn't know electronics when I started, but now we can do our experiments, and I have a skill that will benefit me and my employers for the rest of my career. I'd argue that that's a great use of a PhD, as long as it's not the only thing I'm doing.
Additionally, in my lab and many others, there simply wasn't any tool we could use to answer all the questions we have, so we had to invent them. That said, these are much more bespoke systems, but as we push the envelope of what we need to observe, instrument making and method development are only going to be more critical.
The inflation in the scientific equipment market is also ludicrous. Since I've been in grad school some of the things I have to buy have doubled in price. And it really is pathetic when it's something like a syringe pump - I mean you're rotating a screw with an electric motor to push a piece of plastic. I get that it's a small market and the vendor has to make a living, but this isn't the best way for a civilization to direct its resources.
My name is Sina Booeshaghi a second year PhD student at Caltech and I built and designed the poseidon system (with the help of Eduardo Beltrame, Jase Gehring, Dylan Bannon, and Lior Pachter) in the Pachter Lab.
If you have any questions I'm happy to answer them!
I was needing a fully digital (controlled) microscope stage and stand recently. I was suprised to find a lack of motorized stages compared to other items. It would be nice to have a modular stage so it could be used with different stands. Rotation and height adjustment would be nice as well.
Have you came across anything similar? It seems like it would be trivial to build similar to your other designs.
Ooo this could be a great extension of the current microscope system! The only thing I've found is this https://www.thingiverse.com/thing:2525155 which while not motorized, allows for height adjustments.
Consider making an issue (tagged improvement) on the GitHub page!
I have been thinking about this problem for a while. I would like a cheap, open source, commercially available kit for xyz gantry systems. That's essentially what desktop CNC machines and 3D printers are and people hack them for all kinds of purposes.
Unfortunately my current conclusion is that there is no solution that satisfies all 3 at the moment. Your best bet for building such a system is looking at cheap 3D printers and CNC machines, but it'll always be a little bit of a hack and extra work because you'll have to remove parts (hot end, drill bit..) and adapt something that was not designed with the intent of being entirely re-purposed.
If anyone here knows of a xyz gantry system kit that satisfies all 3 (cheap, open source, commercially available) please let me know.
It would be fairly trivial to build. Basically, it is just a CoreXY 3d printer without the head.
I have modified a small CNC mill/router to be used as a "paste printer" for printing frosting/ceramic (https://youtu.be/XwjnVzfl0wA). Many people in the 3d printing community have done similar things with their printers.
I would recommend something without a bed that moves in the XY direction. CoreXY designs would be great but there may be limitations on the carriage weight. In that case a gantry style cartesian printer would be suitable but with 2 Y motors to carry the X stage without skipping.
Edit: Also, See Openbuilds.com there are many machines that fit what you are looking for.
Gantry style as I use the term describes a machine that has 2 Y rails carrying an X rail. All rails are independently powered so each has a motor/drive and all of the rails are supported above the work area.
There are several 3d printers that use this configuration but not all of them are necessarily the exact design I would use.
Thanks. Any chance of adding useful vendor neutral export formats for the repo? eg IGES or STEP?
STL is alright for 3d printing, but not much else. IGES and STEP can be imported into most cad/cam tools, so should make it useful to a potentially wider audience. :)
Great work! I love seeing these openhardware projects becoming more common, especially in the medical/biomed areas.
So many amazing things spawning directly from the RepRap project and community it makes me feel fortunate to have been a heavily active member of the community for the last 8 years.
Opentrons is another machine you may want to look into. They make a machine for pipetting. They use control boards which come from another project I work very closely with (smoothieware/smoothieboard).
Opensource moves so fast and changes the world so much it cannot be overstated. Freeing the knowledge is important to move our species as a whole into the future. I believe every person in tech should seriously at least consider releasing their work to the world without cost, there are so many advances which can be made if people just have the necessary information and tools.
I actually mentioned the poseidon project to the CPO of Opentrons a few months ago, because it would be great to have these kits commercially available and I thought it was a perfect fit for Opentrons (the company). Understandably he said that at the time they really wanted to focus on developing the OT-2 and the conversation ended there.
That being said the poseidon project is distributed under a BSD 2-Clause license [1] and I'd be happy to see someone bringing it to market.
Oh, this is awesome. I build a lot of random equipment, how do I help? This kind of thing could kickstart a lot of novel systems that'd take forever to design or cost a ton otherwise.
I love your enthusiasm! Why don't you shoot me an email and we can chat some point next week.. It would be really cool to get a great community around the project, extending capabilities and adapting the core technology to random equipment.
I work with Lior at Caltech. He's an amazing guy, probably the only person I know who is capable of doing first-rate work in both biology and mathematics.
33 comments
[ 2.2 ms ] story [ 79.3 ms ] threadhttps://publicdomainchronicle.org/
Public Domain Chronicle combines elements of defensive publication, open-access scholarship, and commons licensing to make public domain advocates the fastest runners in the race to publish, preventing others from patenting their findings.
PDC's disclosure form for findings in biology is shorter and easier than any standard corporate or academic invention-disclosure process, and produces an immediate, public, republishable prior art record.
https://pdc.biobricks.org/publish
It's early days for the project, and we're seeking out as many potential researcher-contributors as possible. We're particularly keen to hear from academic scientists and folks in corporate tech transfer offices who may prefer PDC to expensive defensive publication services.
sobs
Great questions. The target audience for these instruments are scientists and hobbyists who don't want to spend a ton of money on similar commercial systems, and want the flexibility to modify their operation (e.g., poseidon can run custom flow profiles per experiment whereas off-the-shelf commercial systems typically only run one flow rate per experiment). The purpose of poseidon is to show that open source biological instruments can be developed and used by a community, similarly to how open source software tools in biology are developed and used.
Academic budgets vary from institution to institution and are sometimes determined by exogenous forces beyond the lab's control.
A complete commercial system to do single-cell RNA sequencing costs tens of thousands of dollars! Using alternatives such as the Harvard Apparatus syringe pumps and DropSeq [0] to run the same experiment will still cost you into the thousands of dollars. With the poseidon system, we greatly reduce these costs. Users can build the instruments to run these experiments for less than $400 and are not restricted to additional costs and tedious firmware upgrades to expand the system.
In response to your point on time management and instrument-making, I think that if there exists a need to develop these systems such that they will advance biological experiments then it's totally cool to have academics work on these sorts of projects! Biologists and bioengineers have always developed tools alongside discovery and this is no different from developing bioinformatics tools.
[0] https://www.cell.com/abstract/S0092-8674(15)00549-8
The poseidon system was explicitly designed with ease of assembly in mind. If you look at the build videos [1] you'll see that assembly of the entire system (3 pumps + microscope station) takes less than an hour and requires just pliers and screwdrivers.
The importance of ease of assembly was a lesson that we took from assembling the miniDrops microfluidics station [2] developed specifically for one kind of experiment (dropSeq) [3]. The miniDrops is very good at what it does but assembly was somewhat cumbersome: it required ordering a custom PCB, specialized parts only available from one vendor (whom I had to nag over the phone to send me a quote!) and assembly of the device itself took 10-20h.
Not every kind of equipment can be made as easy to source and assemble as we did with poseidon, but we really think that keeping this at the front of your mind can make or break the adoption of a piece of open source hardware. This is especially true in the context of biology laboratories, where many people are not what you could call "hackers" or "makers" and will be immediately put off by a daunting assembly process.
[1] https://pachterlab.github.io/poseidon/hardware
[2] https://metafluidics.org/devices/minidrops/
[3] https://www.nature.com/articles/s41467-017-0265
This is an age old question. And the answer is: yes, we actually do.
Even if a lab can afford all its instruments, sometimes you are researching something that requires a different approach. Most commercially available lab machines are NOT hackable, and you will have a difficult time bending them into a different configuration.
And a lot of the time that equipment isn't funded by a current project and you need to prove the idea is sound before you can justify asking for the money to do it. Of course if that proof of concept is good enough to keep being used all the better.
Additionally, in my lab and many others, there simply wasn't any tool we could use to answer all the questions we have, so we had to invent them. That said, these are much more bespoke systems, but as we push the envelope of what we need to observe, instrument making and method development are only going to be more critical.
The inflation in the scientific equipment market is also ludicrous. Since I've been in grad school some of the things I have to buy have doubled in price. And it really is pathetic when it's something like a syringe pump - I mean you're rotating a screw with an electric motor to push a piece of plastic. I get that it's a small market and the vendor has to make a living, but this isn't the best way for a civilization to direct its resources.
If you have any questions I'm happy to answer them!
Have you came across anything similar? It seems like it would be trivial to build similar to your other designs.
Consider making an issue (tagged improvement) on the GitHub page!
https://github.com/pachterlab/poseidon
Unfortunately my current conclusion is that there is no solution that satisfies all 3 at the moment. Your best bet for building such a system is looking at cheap 3D printers and CNC machines, but it'll always be a little bit of a hack and extra work because you'll have to remove parts (hot end, drill bit..) and adapt something that was not designed with the intent of being entirely re-purposed.
If anyone here knows of a xyz gantry system kit that satisfies all 3 (cheap, open source, commercially available) please let me know.
I have modified a small CNC mill/router to be used as a "paste printer" for printing frosting/ceramic (https://youtu.be/XwjnVzfl0wA). Many people in the 3d printing community have done similar things with their printers.
I would recommend something without a bed that moves in the XY direction. CoreXY designs would be great but there may be limitations on the carriage weight. In that case a gantry style cartesian printer would be suitable but with 2 Y motors to carry the X stage without skipping.
Edit: Also, See Openbuilds.com there are many machines that fit what you are looking for.
I'd like to have a generic open source xyz gantry platform that I could just order on amazon for <$200 and tinker around with.
I used RAMPS to control the steppers.
not clear if it satisfies your open source criterion.
Here is a couple videos of one I made using "PBC Linear SIMO stages" it uses 10 start, 25mm/turn leadscrews so it is a bit louder than normal printers. https://youtu.be/fQ9PoQIl7q4 https://youtu.be/zG-eEfDkFBo
https://github.com/pachterlab/poseidon
STL is alright for 3d printing, but not much else. IGES and STEP can be imported into most cad/cam tools, so should make it useful to a potentially wider audience. :)
So many amazing things spawning directly from the RepRap project and community it makes me feel fortunate to have been a heavily active member of the community for the last 8 years.
Opentrons is another machine you may want to look into. They make a machine for pipetting. They use control boards which come from another project I work very closely with (smoothieware/smoothieboard).
Opensource moves so fast and changes the world so much it cannot be overstated. Freeing the knowledge is important to move our species as a whole into the future. I believe every person in tech should seriously at least consider releasing their work to the world without cost, there are so many advances which can be made if people just have the necessary information and tools.
That being said the poseidon project is distributed under a BSD 2-Clause license [1] and I'd be happy to see someone bringing it to market.
[1] https://github.com/pachterlab/poseidon/blob/release/LICENSE
My email should be in my profile.
My friend André Maia Chagas is doing some inspiring work in open hardware too: https://mobile.twitter.com/chagas_am