How the Broad’s Genomics Platform ramped up human and viral sequencing and kept processing COVID-19 tests, all at the same time

This year has been the biggest yet for the Broad Institute’s Genomics Platform. The team continued its large-scale COVID-19 diagnostic program, launched in March 2020, and is processing up to 120,000 tests each day. In collaboration with the Broad’s Viral Genomics Group, they are also sequencing up to 10,000 SARS-CoV-2 viral genomes a week to look for the omicron variant and monitor other variants of concern.

If that wasn’t enough to keep them busy, the platform massively expanded its existing human genetic sequencing and genotyping analysis capabilities for research and clinical work. Of the more than 340,000 human genomes sequenced at Broad since 2013, more than one-third of those were completed in 2021 alone. The team picked up speed, too. In 2019, they produced a clinical-grade human genome sequence every 12 minutes; now they are doing this every 4 minutes. All told, the platform has sequenced more than one million people, and performed genotyping analysis on 1.4 million people since 2013.

We spoke with the platform’s translational genomics senior director Niall Lennon, an institute scientist at the Broad, and clinical laboratory associate director Tom Howd to find out what’s driving this incredible pace, what’s enabling the effort’s growth while they’re still responding to a global pandemic, and what’s in store for the group down the line (hint: more big years are still to come).

Q: What’s driving the recent increase in human genome sequencing in the platform, and how did you balance that with your response to the COVID-19 pandemic?

Niall Lennon

Lennon: Until recently, we were sequencing around 30,000 to 40,000 human genomes each year, for a variety of projects. When the pandemic hit in early 2020, we’d already spent over a year working furiously to get FDA approval to support All of Us, a research project launched by the National Institutes of Health aimed at sequencing the DNA of one million people in the U.S. to build a diverse health database. We’d been named as one of the genome centers for the project, and were planning to sequence a majority of those samples here at the Broad, in addition to performing genotyping analysis on a smaller number of samples.

When we were awarded approval in the summer of 2020, that unlocked thousands of samples that the NIH had collected over the past couple of years, which were waiting to be sent to us for sequencing. So while the team was building out the COVID testing capacity in our labs, we began receiving a lot more samples for genome sequencing for All of Us, in addition to addressing a backlog of pre-existing requests that had been delayed when we essentially shut down our sequencing efforts from March to June 2020.

Tom Howd

Howd: As we prepared for All of Us, we had a big team effort to think about the systems we already had in place, to see how we could jump up to processing 150,000 or more samples a year. We more than doubled the size of our fleet of sequencing machines, enabling us to generate much more data. And then to prepare all of the samples, we re-envisioned how we operate in the lab, so we can handle that higher scale without having to increase our head count significantly.

I think one of the other challenges was that, at the same time, we were working so hard to increase the scale of the COVID diagnostics, which utilized a lot of the same R&D, personnel, and space resources that we had. We had to improvise and think about how we could scale up sequencing in new ways, in light of the ongoing and arguably more time-sensitive issue of getting the COVID diagnostic lab scaled up.

Q: How did you adjust staffing in the lab and beyond to keep up with all these demands?

Howd: Over the past couple of years, we’ve added to our sequencing staff a bit. As we increased our pace of sequencing four-fold, we’ve only had to increase the number of team members by 50 percent to keep up.

During the early months of the pandemic, our staff members who’d previously been in sequencing operations temporarily pivoted to work in the COVID diagnostic lab. From a regulatory perspective, many of them were already able to work in clinical testing, so they became the early COVID testing staff. As the pace of testing picked up, we worked out three-day on, three-day off shifts, including overnights, and Niall and I worked some of those early shifts too.

It became clear that we’d need the COVID diagnostic operation to continue through the pandemic, but we also needed to resume our previous sequencing work that had been paused. That’s when we began hiring new personnel for the diagnostic lab — eventually hundreds of staff members over the course of the pandemic — and returning those other folks back to sequencing efforts as we began the All of Us work.

Lennon: The COVID-focused teams have been working in this building since March 2020, and it’s never really shut down for any appreciable amount of time. That’s been amazing. None of this would have been possible, even with all the automation we have, without the dedication of these team members.

Howd: It goes beyond the lab, as well, to many cross-functional teams that have enabled this, with folks on the informatics side enabling the transfer of huge amounts of data and return of results to the clients, to project managers and people responding to many questions that came in daily. We’ve also had the support of Broad’s leadership, administrative groups, recruiting teams in human resources, and colleagues in purchasing. I’ve come to realize that everybody seems to feel a sense of ownership and pride over the work that they're doing. They understand the impact of their work, and I think that helps to drive the progress and our continued ability to produce high-quality data, whether it's in the testing sphere or the genomic sequencing world.

Q: Looking back on the history of the Genomics Platform at the Broad, was diagnostic testing or pathogen surveillance part of the vision for the group?

Lennon: We've long been playing in this space. When I joined the Broad’s sequencing group 16 years ago, the very first project I worked on was sequencing of viral genomes for surveillance and mutation-tracking of dengue virus in places where it is endemic. We later worked on HIV, West Nile Virus, and other viruses. We’ve always been active in understanding the technologies available to generate this type of data.

Then seven years ago, we founded the clinical lab at Broad because we did see the community start to move into that space where generating this type of data was moving beyond just basic research. It became clear that being able to return those results to patients would become more and more a part of all the genomics projects. If we find out someone is at risk of developing a disease, we have an ethical responsibility to tell them that, and it’s also good to keep people engaged with the science. In everything we’ve done over the past ten years, we’ve been building towards that. Even when we didn’t have a lot of samples to process, we believed it was going to be important in the future. It turns out we were right.

Q: What will the next few years look like for the Genomics Platform?

Lennon: As we've seen in the last two years, it's very hard to have a crystal ball. Sometimes things will come along and we have to respond to that, and we’ll have to pivot a little bit.

On the genomics side, we’ll continue our translational move from mostly research-grade genomes a couple of years ago into more clinical pipelines today and beyond. We’ll lean into what we’ve built for clinical genomics at scale and collaborate with groups who are working to show the utility of the genome in the clinic, for things like preventative medicine, polygenic risk scores, and diagnostic screening. Advancing the field of clinical genomics to accelerate biomedicine has always been core to our mission. So we’ll continue to serve the Broad community by offering genomes and exomes at scale and at a low cost, so that Broad investigators can keep generating the large data sets they need for discovery in common disease or for diagnostics in rare disease.

We’ll also pursue other smaller scale efforts, such as evaluating new sequencing technologies that might be lower cost or more powerful, in addition to single-cell tools to support studies of cell circuitry and the function of genetic variants in health and disease. All of these are aligned with the institute’s scientific vision, so we want to apply what the Genomics Platform excels at — quality at scale and high-complexity analyses — to support our community members’ existing project or help secure funding for other audacious initiatives.

Regarding our pandemic response, we will serve the community with COVID diagnostic testing and surveillance sequencing to look for emerging variants as long as we're needed. When we get out of this pandemic, we’ll be having deeper conversations about how to keep those capabilities in place, so we can quickly respond to future public health needs.