Life science lab automation today is in the same place computers were in the 1970s: about to make the transition from a niche novelty to mainstream ubiquity. This is the story of how the democratization of computers happened, and how lab automation is going through the same transition today.
The idea of a “personal computer” (PC) made its mainstream debut in 1977, the year the Apple II was launched. Before the Apple II, the main way people interacted with computers was through “time sharing” on centralized “mainframe” machines. Rather than having their own computer in front of them, people would have a terminal that was connected to the central mainframe, and multiple people would share the runtime of one computer. Most people never even saw the actual computer, it was so complicated and arcane that only very specialized engineers were allowed in the room where it was kept.
In those early days, this paradigm worked well for the large corporate and government organizations that used computers to process big jobs like payroll and taxes. It worked so well in fact that Thomas Watson, Chairman and CEO of IBM in the first half of the 20th century, is said to have stated that, “I think there is a world market for maybe five computers.” The idea that anyone outside of a massive organization would need to use a computer — much less own one (or many!) themselves — was seen as impossible and undesirable.
A typical mainframe style computer room
Today, this is much the same view many hold of lab automation. The lab robots of today are true “mainframe” style machines — big, expensive, and requiring expert automation engineers to run them in central facilities. They are great for batch processing high-throughput experiments, but remain inaccessible to 90% of labs in the world.
A typical mainframe style lab automation facility
For the biologists lucky enough to have access to lab automation, usually it is only in a “timesharing” style harking back to centralized mainframes. Whether you’re at an academic lab or in industry, usually researchers access automation through their institution’s “core labs,” rather than operating it themselves. And, like operators of the mainframes of yore, core labs will split up work given to them by researchers and do “batch processing” to divide the core’s automation capabilities efficiently between jobs. In a few days (or weeks, if you’re less lucky and/or well funded) after you submit your job, you’ll get your data back from the core lab.
Opentrons is on a mission to change this paradigm. While mainframes are great for very high-throughput work, they should not be the only option available for automating lab work. Scientists should be able to automate their own work. We need a personal lab robot, a “PC of lab automation,” so people can directly experience the accelerated workflow lab robotics provides. That’s why we made the OT-2, the world’s most affordable high-precision lab robot.
There are three keys we see to making a truly “personal” lab automation platform.
The OT-2 fits on half a lab bench
The OT-2 hits all these marks. Starting at $5,000, it is priced below most lab’s discretionary spending budget, meaning they don’t even need bureaucratic approval to purchase their robot in most cases. We’ve incorporated the user experience (UX) processes of consumer technology companies, holding ourselves to the same ease-of-use standards followed by Apple and Google. And our machine is the only lab automation platform on the market that is both fully modular — making it easy to add or remove features as needed for researchers’ workflows — and allows people to share their automated protocols easily with collaborators.
But just because we put personal lab automation on the market doesn’t mean it is going become a workhorse in every life-science lab. The path of computers scaling from obscurity to ubiquity did not happen because everyone is a nerd and wanted to play around with BASIC programming; it happened because of the powerful things people could use their PC to do. This is the idea of a “killer app” — the application of a technology that causes mainstream people to use it (rather than early-adopter tech lovers). For the Apple II, the killer app was VisiCalc, the first spreadsheet program. With VisiCalc installed, the Apple II became a game-changing tool for businesses to manage their finances. All of a sudden, the world shifted; instead of only the very largest businesses needing computers, every business needed a computer.
This is why Opentrons Protocol Library is so important. It gives people access to applications that are ready to be downloaded and run on their OT-2 lab robot. These applications — PCR setup, NGS library prep, nucleic acid extraction, serial dilutions, and many more — are done in life-science labs around the world on a daily or weekly basis, and can take hours each time a scientist runs them by hand. What’s more, these applications are developed along with our community of users and partners, meaning they are “for scientists, by scientists.”
It is still the very early days of personal lab automation, but already we’re starting to see how it changes the way people do their lab work. Keoni Gandall at Stanford has been able to produce 100x more genetic designs every week with his OT-2; the Aiden Lab at Baylor College of Medicine is able to walk away from their NGS library preps and focus on data analysis, rather than spend a majority of their time pipetting; researchers at the Chan-Zuckerberg Biohub have automated tedious cell culture steps so their researchers never have to worry about manual trypsinization again — the list goes on and on. Scientists in 45 countries around the world ranging from big pharma to tiny startups, MIT and Stanford to community colleges, are multiplying their productivity and moving faster towards their scientific goals than they ever thought possible.
Where will we be in five or ten years when every scientist has their own lab robot, just like we all have our own computer? Just as the rise of personal computers changed what was possible for information processing and media, personal lab automation will change what is possible in life-science. What will be the difference in our global bandwidth for life-science research when that happens? Will we be 10x faster to a new discovery? 100x? 1000x? What will we create with ubiquitous automation that today is too far away to even dream about? I for one could not be more excited to see the world we can make — all the amazing cures, products, and ecosystems we can create — when biologists are finally freed from monotonous pipetting and given the tools to create their dreams.
