Laboratories of the future: automation and machine learning

Discover how a pioneering lab is leading the way in automation and what this means for the future of research

Laboratory automation is the future of research. Pioneering labs leveraging state-of-the-art, fully automated closed-loop systems and integrating machine learning demonstrate the possibilities it can bring.

In this article, we speak with Dr. Stephan Lane, Biofoundry Manager of the Illinois Biofoundry for Advanced Biomanufacturing (iBioFAB) at the University of Illinois, Urbana-Champaign about the iBioFAB team’s approach to automation, what they’ve learned since the biofoundry’s inception, and what it means to us can tell about the future of laboratory research.

The iBioFAB 2.0

Founded in 2014 by the Huimin Zhao Laboratory, the iBioFAB “aims to accelerate the biological engineering process by integrating artificial intelligence/machine learning with automation.”¹ This includes an articulated robotic arm that runs along a track and transfers samples within microplates to instruments. “Our very broad goal is to automate synthetic biology,” Lane explains.

After several years, the biofoundry underwent an upgrade, leveraging knowledge and technological advances to improve the use of automation and introduce new features to the laboratory. This is the iBioFAB 2.0.

The iBioFAB 2.0 uses many different automation tools and software applications to enable automation throughout the lab. Making automation the core of their lab’s function has changed the way their lab works and the kind of work they can do. As Lane explains: “[Automation] opened a door to many possibilities that weren’t there before, and it changed the status quo for the lab. Back in the day, if a student was doing a thesis project and they were developing 20 cell lines and then characterizing them, that was the norm. With the automation, their 20 is no longer enough – now they need 200 to meet the set standard. High throughput has become normal rather than the exception as before.”

Lane explains the ethos of the lab and what the goals were with the upgrade: “We’re trying to do proof-of-concept work, things that haven’t been done before. So when we decided to upgrade, we wanted to add as many features as possible to enable those proof-of-concepts in new areas. Actually, when we thought about upgrading, we were trying to manually think through all the techniques we use in the lab, and which of these we could automate in a high-throughput way. Our goal was to create as many workflows as possible. How do we tackle the things we do in the lab and make them automated and with high throughput?”

“The kind of skills you need are different from a standard synthetic biology lab when you move to automation,” Lane continues. “So we now have an automation engineer, an electrical engineer and a software developer. These are skills not normally found in a synthetic biology lab. We don’t have tunnel vision and we try to use the skills we have and deploy them when they become useful.”

Connectivity is key

An essential aspect of an automated laboratory is the connectivity between different operations. In iBioFAB 2.0 they use closed-loop systems to connect individual functions within a workflow and link them together via software. Connectivity in these closed systems allows automated workflows and experiments to be performed with high throughput.

Because high-throughput experiments are the future of lab work, labs also need a way to analyze the large amounts of data produced. To support this, machine learning or artificial intelligence algorithms can be incorporated into the closed system to enable a complete workflow that includes data analysis.

Lane gives an example of how a closed system in the iBioFAB lab works to generate and analyze large amounts of data: “We had data from our analysis tool, in this case a plate reader, fed directly into our machine learning algorithm, which was in Python. Using the … Thermo Scientific™ Momentum™ Workflow Planning Software, we can automatically take the data output and enter it into a Python script. By linking this to our inventory of different pieces of DNA, the algorithm can select which ones will be of most interest, and we can reassemble those and start the process all over again. Connectivity was key to that and everything we did was coded into our Momentum software workflows as well.”

Machine learning or artificial intelligence embedded in these closed loops is essential to both understand the data and inform the experiments to be performed in the first place. “A good example of this would be the BioAutomata, the closed loop paper we published²‘ explains Lane. “This kind of work described about 14,000 possible assembly combinations that could have been explored. And trying to do something like that without automation will be impossible. Even if you did that with automation, it would take weeks and weeks just because of the throughput limitations. By combining it with machine learning, we can take the best of both worlds and explore not every possibility, but a subset, so we can narrow down to one of the best performing.”

What does the future hold for laboratory automation and access for researchers to the tools and advances that automated laboratories can provide? “I foresee that a place like the one we have with our biofoundry will become commonplace in every university,” concludes Lane. “A facility where researchers can hire time, or staff to conduct high-throughput studies, because it’s not feasible to have a huge robot for every lab, and it would probably be a bit wasteful to do that. to do. So probably in the future there will be a centralized facility that can handle these high-throughput assays.”

References: