Kevin McKernan
speaker
709 appearances
2 recordings
2 series
first heard Nov 2024
last heard 5 Dec
Kevin McKernan’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsRecordings per month over the last 12 months — 1 in all, peaking in Dec 2025 with 1.
Appearances
So we got in this big race with Craig Venter to sequence the human genome as quickly as possible. Right, right, right. I was on the government side. And then when that came to a close, I ended up dropping out of a PhD program and starting a company.
We started this company called Agincourt, which took a lot of the technologies we built in the genome project and commercialized them and started selling them to every biotech company that wanted to catch up, if you will, to what they were doing down at MIT to sequence this quickly.
So that company, a few years later, Francis Collins funded it for an obscene amount of money to be one of the five genome centers in the country. I think we were probably the only private one that was in there. It was Eric Lander at the Whitehead had one of the grants. Baylor had one of the grants with Richard Gibbs, who's a hilarious guy. You should have him on one day. And then there was St.
Louis, Wash U, St. Louis had a genome center, ours, and then Venture they also funded. So we ended up becoming a sequencing shop, specializing in that. And in the course of doing that, we decided to just change sequencing, like reinvent the whole thing and build a new sequencer. Also received a grant for that.
So the Agicorp company got acquired in 2005 by Beckman Coulter for the genetics pipeline and all the magnetic bead tools we built to purify DNA. There are these little magnetic particles you can use to capture DNA and rip them out of solution. It's really handy for studying viruses. And everything else. And that was kind of the cornerstone of the IP that we had at Agincourt.
But when that happened, we also had played around with taking those beads and putting them down on glass slides under microscopes and sequencing DNA individually off of every single bead. So that we, instead of doing 96 things at a time, we're doing 100 million things at a time. It was a massively, just totally different way of doing it.
We couldn't get the read lengths that we're used to getting, which we used to be able to read like 600 bases or so with these older tools. But this thing, you can get like maybe 30 to 50 bases, but you can get 100 million of them at once in the same timeframe. So it was a total game changer in the sequencing field. Ultimately, we were getting genomes.
They used to cost about $300 million to do the first human genome. And this thing was pumping them out at like $3,000. Whoa. And so once we had proof of principle of that working – We had to spin out a company, and this 19-person company got in a bidding war with Lumina and ABI, where we're the two largest genetic companies in the world, started having a bidding war over this company.
And they eventually acquired that company, and I went with that acquisition, and we started building those. These things were called solid sequences. We started putting them to the market, and the first people that were picking those up What were they called? What kind of sequencers? Solid. Solid sequencers. Yeah. Standard for sequencing by oligo ligation detection.
We did sequencing a different way. I mean, most people use polymerases that grow DNA strands with enzymes on – in one direction. If you remember your biology, you have two DNA strands. And when they split, polymerases copy one strand. The other strand is replicated with something known as lagging strand synthesis. It's a different way of replicating DNA that your cells use.
So one gets done with polymerases. The other side gets done with ligases. All the IP in the world was covered on the polymerase front. That's what everyone had been using since Fred Sanger. And so we realized that that's just a minefield of patent estate. And like, we're going to go on the other strand and do it backwards.
And so we started doing these ligation things and it worked, worked really well. It's still to date, I think has the record as, as probably the most accurate sequencer ever built, but it's just limited in the length of reads they can do. It can probably only do like 50 to a hundred base pair reads. And now these other sequencers that are out there can do thousands to a million base pair reads. And,
They take a hit on some of the accuracy of the sequence, but they can get lengths that we could never get on the system that we had back then. But nevertheless, it was a powerful tool. It got distributed to lots of cancer centers that were sequencing tumors to try and differentiate tumors from the patient's genome.
And this whole personalized medicine thing was really looking like it was about to shine. So I was there for about five years.
Oh, they initially were going after – there was a group, Mike Stanton's group, I think, in Sanger Center used this to look at lung cancer. Glioblastoma got sequenced with it down in UCLA. And there was a group down at Baylor looking at breast cancer. I mean, it got used across the spectrum of different cancers, colon cancer.
I was on a paper with a group at Johns Hopkins. They're on top of the game with all the stuff down there at Bert Vogelstein's lab. And they were doing something very interesting with it in that.
Cancer, when you have it, oftentimes is sloughing dead cells out into your bloodstream. And if you sequence your bloodstream, you can kind of track which cells are dying from the cancer and look at their mutations and try to get a profile of whether the tumor is getting better or getting worse by just non-invasively sequencing bloodstreams.
So Rebecca Larry put out this paper where they did that, and we're scanning – people over time through the course of treatment by sequencing their bloodstream over time and developing markers that were very personalized to their tumor that would tell them that your particular tumor is going up or down based on what we're doing.
And the same tool got rolled into amino – sorry, I thought a little hot.
Amniocentesis. So the other thing circulating in mother's bloodstreams, around 6% of the DNA in a maternal bloodstream is actually the baby's. So if you don't want to do amnio, something that I'm sure you're familiar with given your recent father, amniocentesis has like a 1 in 400 fatality rate.
Showing 301–320 of 709 · page 16 of 36
← Previous
Next →