An underappreciated truth about drug discovery and exploration in the life sciences: They take a lot of hard work. Even the basic chore of cultivating the right types of cells or tissue for research can require months of repetitive, painstaking effort.

That's about to change - at least for researchers at UCLA and beyond. A fully automated robotic system for growing and experimenting with cells and tissue is now in operation at the Molecular Screening Shared Resource, led by Robert Damoiseaux. The MSSR is a technology center located at the California NanoSystems Institute at UCLA, or CNSI. The new system, called the CellXpress.AI and made by Molecular Devices, performs the full range of handling liquids, incubating cells, capturing images and analyzing data. Robotic components work precisely and autonomously, while artificial intelligence manages steps, watches over cultures, analyzes images and crunches numbers.

Funded by a $1.9 million High Impact Equipment grant from the National Institutes of Health, the Cell Xpress.AI at UCLA is the only device at a public research institution on the West Coast. The instrument will be available for campus-based collaborations, as well as collaboration with other universities and partners in industry.

"The bigger context for the new instrument is making science more accessible, credible, scalable and reproducible," said Damoiseaux, the principal investigator on the NIH grant and a UCLA professor of molecular and medical pharmacology and of bioengineering. "These are all things we need if we're going to ensure resources are used wisely - including precious taxpayer money."

The Cell Xpress.AI is especially well-suited for turning stem cells into organoids, cell systems that mimic organs in the human body. Studies in organoids can manifest better models for human biology than experiments in animals, as they can be grown from human cells.

How the Cell Xpress.AI can be used by researchers

For example, in drug development, researchers can test a candidate compound with kidney and liver organoids to assess toxicity. Organoids are also vital for advancing knowledge of topics such as the development of the brain and immune system, the wiring of circuits within the brain, and health conditions including cancer, lung disease, dementia and neuropsychiatric illness.

However, growing organoids in the lab takes months and involves daily, labor-intensive tasks.

"Systems that give you predictive insight without animal models have been at the forefront of our minds for a while now," Damoiseaux said. "Fortunately, science never sleeps, and we're at a point where we can make organoids through advanced methods in stem cell culture. Now, we can also take the manual labor and variability out of it."

The CellXpress.AI creates a new level of flexibility for investigators in the life sciences. A microscope with a 21-megapixel camera feeds into a type of AI patterned after the workings of the brain, which assesses cells without the need for staining or fluorescent tracers. This automated imaging guides the experiments underway and generates rich data. And adjustments based on the analysis can be made midstream.

"Any good experimenter with cell culture experience can look at their cells, see the state they're in and identify changes they would make to the media or other experimental parameters," Damoiseaux said. "That's something that we can do with this system. People in our business currently talk about design-test-analyze cycles, but this is slow. With the new system, we are transitioning to 'design, test and analyze at the same time, then modify on the fly.'"

Tech expertise not required

The MSSR team provides the know-how to realize the most complex experimental workflows - enabled in part by their longtime collaboration with the company behind the instrument.

"We can get as complicated or simplistic as we need to," Damoiseaux said. "In addition, the end user can tap into the power of AI without having to be a mathematician. You can train the system to your specifications, so it doesn't negate the individuality of your technique. But we also have experts like myself who help people delve into designs and workflows."

Conducting studies with the labor of robots benefits both the science and the scientists. The work gets done more efficiently, the same way each time. Results become more reliable.

"Whenever you put a human being in front of a system, it's going to be slightly different because people have slightly different techniques," Damoiseaux said. "We're taking all of these variables out so your end product is uniform."

With tedious work offloaded, researchers can channel time and effort into higher-order tasks beyond the ability of machines. Damoiseaux emphasizes researchers can look to re-prioritizing, rather than replacing, the people behind the science.

"People worry that AI is coming for their jobs, but that's not really the case," he said. "AI is only coming for a repetitive job, the job you don't really want to do. Instead, you can be conceptualizing new ideas or analyzing your data.

"After all, the system can't do your thinking for you. AI doesn't ask good questions. It's not very inventive."

With MSSR's Cell Xpress.AI brought online only recently, these are the early days of autonomous experimentation in UCLA bioresearch. The device opens up unprecedented opportunities for faculty across disciplines at CNSI, the David Geffen School of Medicine at UCLA, the UCLA College's divisions of physical and life sciences and the UCLA Samueli School of Engineering. Damoiseaux envisions that automation will continue to transform the way investigations are done.

"This instrument is going to make it a lot cheaper to generate solutions," he said. "It will enable advanced cell and organoid models that we can't generate at scale otherwise.

"At the same time, this is just the first step toward a lab that's more autonomous. Remote experimentation is where this is heading, where your data is getting generated in the lab and you're analyzing it at home. Right now, we're at a good starting point."