There aren’t enough tests. That’s the problem now on many of our minds.
But as the federal government fails to meet demand, local institutions are stepping up. The latest lab, opening at UC Berkeley next week, will churn out diagnoses in 24 hours. Its workers are robots: industrial drones like the ones that build cars and computers, but for lab-tech work. These boxy bots are miniaturized factories, with all the steps automated. They lift test tubes, drip doses of chemicals, and haul samples into neat rows, each branded with a barcode. In their day jobs, the robots pull genetic material from cells using magnetized “hands,” to study drugs, profile diseases, or learn how cancer develops. Now they’ll moonlight as testers in the war on the virus: running up to 4,000 tests each day.
The bottleneck to slowing the virus is not medical, but material. Currently, the average test for the coronavirus takes four to seven days. Manufacturers can’t produce, package, and ship enough test kits to meet the growing need. Overall, the U.S. is testing at a rate 3.5 times slower than South Korea and well behind Canada. Over 3,000 Americans have died of coronavirus, more than were killed on 9/11. Roughly 180,000 are infected, that we know of. But this number is vague. We can’t test enough to know the actual figure.
The scientists at Berkeley are part of the Innovative Genomics Institute (IGI), which was founded by biochemist and CRISPR pioneer Jennifer Doudna in 2014. The mission of the institute is to empower medicine and agriculture with genetic engineering, linking doctors and biologists at UCSF and Berkeley. The idea for the testing facility was born just two weeks ago, when a virologist in Doudna’s lab saw her own projects put on hold.
“I was talking to Doudna when we were shutting the lab down [for quarantine],” recalls Jennifer Hamilton, who’d studied influenza at Mt. Sinai before coming to Berkeley. “I told her I was feeling a little bit left out because I wasn’t able to apply my training.” Doudna encouraged Hamilton to get a group together, and the result was swift: 800 researchers volunteered, and soon they had a speedy coronavirus test.
“What’s really struck me about this is the interest of people wanting to volunteer and help in this process,” says Hamilton. “It’s been remarkable what we’ve been able to do in two weeks.”
The tests are powered by those robotic “lab techs,” which were re-purposed from their usual role in industrial-scale genetic experiments to detect coronavirus on a mouth swab.
“The robot is about the size of a lab bench,” Hamilton says. It’s called a Hamilton Vantage (no relation) and it moves liquids using magnetic plates. “So the robot can do all these RNA extractions that involve adding liquids and taking liquids away.”
The new testing facility at Berkeley is a reaction to government neglect. In The Atlantic, Ed Yong called the testing fiasco “the original sin of America’s pandemic
The bots do the dullest, most essential but procedural work of chemistry: moving and mixing the chemical ingredients that make reactions happen––from breaking down substances with enzymes to reading out DNA. Scientists use a device that looks like a tablet to program the bot, commanding it to execute routines, or assays. A light on the box’s top shines green or red to tell human handlers when help is needed, but otherwise the machine is self-sufficient. The bot moves plastic trays, like advanced Petri dishes. Barcodes track the samples’ IDs, linking each well and test tube to an anonymous patient. As the trays enter the central chamber, the “pipettor” arm descends. Black pipettes dose each sample with the needed reagents to prep genetic material for sequencing DNA. A human lab tech with a pipette can drip just one careful drop at a time, as if watering hundreds of tiny plants. The robot can dose 96 samples at once. In this way, the bot army hunts the virus at industrial scale.
A doctor will get the results for each patient. In the beginning, these doctors will all be at the Tang Center, UC Berkeley’s medical facility, but the hope is to expand to other medical clinics in the East Bay. The patient samples will be anonymized using a computerized tracking system and labeled with an identifying barcode.
The new testing facility at Berkeley is a reaction to government neglect. In The Atlantic, Ed Yong called the testing fiasco “the original sin of America’s pandemic failure.” What happened was this: The U.S. chose from the beginning, as it typically does, to make its own coronavirus test, rather than use the test made by the World Health Organization. On February 5, the CDC sent its test to 100 labs across the U.S. But the test failed. In the meantime, a tsunami of virus swept across the states, leaving governors and mayors to make health decisions the president failed to make.
Frustrated by the government’s failure to meet the pandemic’s needs, doctors and scientists around the U.S. have started opening private testing labs. Data miners and journalists have also joined the charge, filling other gaps left by the federal government. (There is no U.S. agency tracking data on coronavirus testing, for example, so a group of volunteers and journalists built the COVID Tracking Project.)
The IGI group hopes to apply CRISPR, the new gene-editing technique pioneered by Doudna, to diagnose viruses and other diseases. Mammoth Biosciences, which was co-founded by Doudna and CEO Trevor Martin, is developing a CRISPR-powered toolkit for other labs to make their own diagnostic tests.
The bots and biologists at Berkeley, meanwhile, are fast closing in on the virus.
“Imagine setting that up—a process that would normally take months to years—in a couple of weeks,” Doudna said of the new virus-testing factory in a press release from IGI. “It’s really extraordinary and not something I’ve ever seen in my career.”