NARRATOR: Luminaries– talking to the brightest minds in tech.
MICHAEL DELL: And my hope is that we come together, to share more than technology and expertise and products, but that we share a vision of a future that is better than today, a vision of technology as the driver of human progress.
NARRATOR: Your hosts are Mark Schaefer and Douglas Karr.
MARK SCHAEFER: Welcome, everyone, to another episode of Luminaries, where we talk to the brightest minds in tech, and we are going to deliver that today. This is Mark Schaefer with my co-host Doug Karr. How are you, Doug?
DOUGLAS KARR: Overwhelmed with this topic today. This is incredible.
MARK SCHAEFER: It’s going to be lots of fun. And the reason I’m so excited about this is our guest is James Lowey. James works for a company called TGen, the Translational Genomics Research Institute. We’ll be talking more about that.
But one of the reasons I’m so excited about this is we’ve talked to a number of different technical leaders from around the world, technological thought leaders. And whenever we have asked them, what are you most excited about? What is the technology that’s really making you dream and be hopeful? They mention medical technology and things that are going on with the genome. So this is going to be an amazing opportunity for us today. James, welcome to our show.
JAMES LOWEY: Hi. Yeah, it’s great to be here. I don’t know about being one of the smartest people, in IT, but I’ve certainly spent many years doing IT.
MARK SCHAEFER: James, I’m just going to read a little bit from what I got off the website about what is TGen to introduce this to our guests. It’s a non-profit medical research institute working to unravel the genetic components of common and complex diseases. So this is just amazing. What can you tell us about the work of TGen today?
JAMES LOWEY: I think that that’s a pretty good description. But the way I would characterize it is it’s really around trying to accelerate research to bring better treatments to patients faster. So I’ve been with TGen for a little bit over 15 years, now, and it’s been an amazing journey as the technology that’s been developed to sequence our genome has evolved.
We went from a period of being able to only sequence relatively small amounts of your DNA, to today where for less than $10,000, you can sequence your entire genome at a very high coverage depth, which really can help us give insight into how we tech. And it’s been an amazing journey.
TGen was formed as the biomedical cornerstone of genomics research in Phoenix, Arizona in 2002. And we’ve been very fortunate to be lead by Dr. Jeffrey Trent, who’s our president and scientific director, who actually came from NIH. And he was one of the original guys to sequence the human genome for the first time. So when you’re working with that caliber of person, it really, I think, lifts everybody up to excel. So I’ve been very fortunate to be a part of that.
MARK SCHAEFER: And the work that you’re doing is so broad. It’s just touching. So many types of people and so many diseases.
JAMES LOWEY: Oh, absolutely. We’re focused primarily around cancer, but we also work with Alzheimer’s, diabetes. We have a center for rare childhood disorders, which are a lot of mitochondrial diseases that are passed through familial– basically inheritance. And the other passion that Dr. Trent has– and I think we share at the Institute– is around underserved populations. So folks that maybe pharma won’t pay as much attention to, simply because there aren’t enough of them. But they need help, too. And the fact is is what we can glean from sequencing these folks will help everybody in the end.
MARK SCHAEFER: And how did you get into this interesting field and this interesting career? Tell us a little bit about your career path.
JAMES LOWEY: Oh, yeah. So it’s been interesting, for sure. I’ve been doing IT for about– little over 25 years, now. I started off working at a university running the computer labs with a bunch of students. And basically the deal was they didn’t pay me much, but I got a free college education. Only had to work about 80 hours a week.
JAMES LOWEY: So that was fun.
DOUGLAS KARR: Free.
JAMES LOWEY: Yeah, exactly. And then from there, I progressed through to various different Fortune 500, 100, 50 companies building out large-scale enterprise systems. I was brought in to TGen to help them build a high-performance computing infrastructure, one of the first major high-performance computing infrastructures in the state Arizona– at least one that people knew about, anyway.
And I’ve been fortunate enough, in my tenure at TGen, to build two systems in the top 500 supercomputers in the world. So it’s been a thing where the amount of data that’s generated by our DNA is truly mind-bending. If you think about it, on average, a patient’s DNA requires about 4 terabytes of storage through the process.
So if you’re sequencing 10 people that, that’s not a huge amount. But if you’re sequencing 10,000, now we’re talking about real data– and so multi-petabyte type environment. So really, the history that prepared me for this was in the high-performance computing and data wrangling at a scale that I think is coming towards everybody. We’re just one of the leaders in that space, I think.
MARK SCHAEFER: Well, and to better define that data, I had read online that a person writing 60 words a minute, 8 hours a day would take 50 years to write one human genome, and the stack of paper would be as tall as the Statue of Liberty.
JAMES LOWEY: Yeah, that sounds about right.
Of course, if I was writing it, it’d be messed up, because nobody could read my writing. That’s one of the reasons I got into computers, because–
DOUGLAS KARR: What is different about building a world-class supercomputer infrastructure that you have to work with that other companies typically don’t?
JAMES LOWEY: Yeah, that’s a really good question. Typically, what you’d see at a university in the supercomputer environment has to support a very broad spectrum of programs– so from astrophysics through engineering and through financial modeling. And so those systems tend to be built as a multi-purpose tool.
What we’ve done, working with Dell EMC, has been to actually build a purpose-driven, high-performance computing infrastructure that specifically was engineered to process data that’s coming off the sequencers. And what this basically allowed us to do is take a timeline of a couple weeks and shrink it down to about eight hours, just through engineering the system. And it’s funny that we did this– designed this system on a white board in my office, and about a year later, Dell EMC actually productized it, and you can actually go to them and buy one of these machines today.
DOUGLAS KARR: Wow, that is fascinating. That’s amazing. That’s a sign of amazing collaboration.
MARK SCHAEFER: Yeah.
DOUGLAS KARR: It really is. That’s so cool.
JAMES LOWEY: Yeah. I mean, this is why– in IT, you generally flip around between jobs every– three to five years is pretty typical. I’ve been there 15 years. I don’t plan on leaving, because I love what I do, and it’s always interesting. The fact is we’re challenged by the science.
The science was moving faster than the IT technologies, so to keep up requires a huge investment in time and effort in understanding new things that are coming. And that’s really where Dell Technologies has been huge, because we get exposed to early access technology, and it helps us build systems that are going to do the next generation of sequencing as this moves from more research-driven into clinical application at scale. So we’re talking about being able to sequence whole hospitals’ worth of patients to help get better treatments to all those patients, instead of just a select few, which it has been in the past.
MARK SCHAEFER: But that’s going to just take massive computational power. So I want to hear more about this idea that you’re talking about, because what fascinates me is that this is truly a technology-based company that could never have existed without computational power. You are centered with software and hardware. That’s really how you can only exist.
And you’ve mentioned this idea of the medicine is outpacing the technology. So interesting. So what are you looking at to bring that back up to speed? What are some of the ideas, some of the technologies that are emerging that you’re saying, OK, we need to pay attention to this, because we got to catch up to the medicine?
JAMES LOWEY: Yeah, no, that’s a great point. And that’s why the theme around digital transformation really resonates, because we’re living that. We have to live that. So we’re looking at, basically, building out hybrid cloud environments that will allow us to transfer workload seamlessly between on-prem off-prem with collaborators so that more people can derive benefit from the work that we’re doing.
We believe in having an open ecosystem around genomics, because at the end, it’s going to benefit all of us. And I think that that’s been one of my passions. Personally, that’s why I spend a lot of time doing things like this, is to help people understand so that more people can go out and do these things, because I think it’s really important.
DOUGLAS KARR: I think that’s the most unique thing that I found about your company– in a technology forthcoming like this that could be highly profitable for the future, you guys basically have created a nonprofit and are giving that research away. How does that change the culture internal and the business processes that you guys work with?
JAMES LOWEY: That’s really a good point. I’m really blessed to have a team with me that’s– most of the folks have been there for 10 years or longer. And I think, to answer your question, that’s how it impacts– is keeping people around and having people engaged. Having a sense of purpose, I think, is absolutely essential to success here, because the fact is, the skill sets that are required to do these things are really hard to find.
And I’m fortunate enough that some of the folks I have working with me are people who I would say are experts and leaders in IT, people who can– you could drop them in any company, I’m absolutely sure. And if they’re listening, that’s not an invitation to leave.
But they will be successful, because that is the kind of people that are required. And I think that really, the culture always comes from the top. And if you ever had the opportunity to meet Dr. Jeff Trent, you would understand that he is driven by helping people. And that just goes downward throughout the entire organization, and it makes it a lot easier to get up and go to work in the morning when you have a mission like that.
MARK SCHAEFER: I love that. I mean, it’s a purpose-driven company, purpose-driven people, purpose-driven technology. And I’m sure you’ve seen a lot of really inspirational success stories that just make you keep going, and that inspires your people as well, which is why you’ve had such commitment and such low turnover. Are there some stories you can tell us about some of the things that your technology is achieving that just makes you go, wow?
JAMES LOWEY: Yeah, it’s probably one of the most amazing things that’s happened while I’ve been working at TGen– was meeting a kid who was about the same age as my daughter at the time who was recovering from a brain cancer. And he was really interested in video games and stuff, so he wanted to meet the guy who ran the big computers, because it was interesting to him. So absolutely.
MARK SCHAEFER: You can’t.
JAMES LOWEY: No. I think it’s really important to remain focused on who we’re trying to help and what we’re trying to do. And I’m a geek as much as anybody. I love computers, and I spend a lot of time with computers. But that’s a bonus to actually having something like that happen.
DOUGLAS KARR: How do you go from that question back to tech? I’m going to try, OK? You talked before about the off-prem, on-prem hybrid cloud solutions that you guys have. I know that when I was reading the research about you, that you let a really key effort to make your data storage more efficient and more effective with all-flash storage. Can you talk specifically about that project and what that did from a business outcome standpoint?
JAMES LOWEY: Yeah, absolutely. We’ve been a longtime Isilon customer. The scale-out capabilities of Isilon really fit with our need to build a storage system that was easily maintainable, was easy to manage, and could grow as we grew.
And with the introduction of the F800 flash, what that did is it enabled us to basically consolidate our storage infrastructure from having a parallel file system in our high-performance computing environment, and moving our entire high-performance computing environment over to a Luster– I mean, not Luster– an Isilon-based solution. And what this bought us is a few things.
One is we were having to– previously, we had to migrate data between storage systems so that bioinformaticians would actually have to spend time doing that, which is probably not a good use of their time. We’d rather have them trying to figure out what was going on with the genomes than being a data shepherd. And so the Isilon with the automated tiering system allows us to take the data from the sequence, or through the entire process, to the variance, which are the files that basically tell you what’s abnormal about the genome in a more automated process, without a huge amount of human involvement.
So really, there’s a huge impact. So now we’ve freed up people’s time. We’ve made it easier for folks to access the data. And it really is all about having that flash tier available, cause that gives us the performance that we needed, cause in the early stages of processing, you’re taking, basically, a 3 billion piece jigsaw puzzle that was thrown into a tornado and trying to put it back together again. And so yeah, it’s really compute-intensive, very IOPS-driven.
It basically is the benchmark that we use to test systems with, because it is just– it’s a really nasty kind of problem. But yeah, the flash stuff– when we benchmarked it, we found that it was totally comparable, if not faster, than the parallel file system that it replaced.
DOUGLAS KARR: Wow.
JAMES LOWEY: That was definitely cool. I remember seeing when that was announced a couple years ago, and it’s like, man. I asked Michael Dell, actually, personally– I said, hey, can you get me one of these to test? And he delivered, and we actually had one of the alpha units that we were able to benchmark, which enabled us to move into the new architecture.
And what’s really cool about that– that new architecture– is that it’s going to give us capabilities that we wouldn’t have had previously, especially around big data methodologies and running Hadoop-based frameworks, because it has native HDFS support on the Isilon. So that really gives us a capability to start looking at data in a new way using tools like Spark, for instance, which there’s a lot of interest in, especially because we’re starting to invest a lot of time and energy in artificial intelligence machine learning algorithms, because we’ve been doing the genomic data for a bunch of years, now.
Now with our affiliation with City of Hope, we’re talking about tying in the data from the EMR systems, as well as other data systems. So we’re talking about building out data lakes that are going to have a really disparate group of data. So we’re going to have to figure out a way to intelligently process this data in order to achieve some meaningful results. It’s a non-trivial task, especially at petascale.
So we have to have the computers do some of the heavy lifting, and we believe that some of the artificial intelligence and neural network technology is going to really help us achieve that. So that’s another thing we’re really excited about. And I’ve heard a few things here at Dell Technologies world so far that have encouraged me that this is something that is a focus, and there is a bunch of energy and time being spent on. And we’re going to derive benefit, and when I say we, I mean the big we.
DOUGLAS KARR: Right, right.
MARK SCHAEFER: Let me spin this another way, or talk about artificial intelligence another way. The obvious benefits are going to be the amazing amount of help it’s going to provide to recognizing patterns and helping with the analysis. I saw an article, actually, just this week that there was some application where artificial intelligence– an AI-driven medical test– it no longer requires human interpretation.
So that was really cool, and it was also a little chilling in a way. So as we go forward, how do you see the human machine interface? How do you see artificial intelligence evolving in the medical field?
JAMES LOWEY: Yeah, I mean, that’s– I read that same article. I know exactly what you’re talking about. Yeah, it’s really interesting, because I believe– and this is just my personal belief– that you still have to have the human subject matter expert to validate what the machine gives you. I don’t think we have a degree of confidence in that result yet.
I mean, to me, it’s always the test. If it was my kid and that was their test result, what would I want done? And today, I would say I would still want a qualified professional to analyze that data.
Now, where I see the value is that maybe instead of having to start with a huge amount of data that that professional has to go through– taking a lot of time and energy, can only do very few patients– to condensing that down to something where it’s pretty easy to say yes or no based upon the data that’s given them, and they can spend less time analyzing data and more time just making sure that it’s correct.
Now, in the long term, what that’s going to look like– I don’t know. I mean, we already have robots performing surgery today. Who knows how far that that can go, to the point where you’re scanned and basically a doctor doesn’t have any kind of intervention, the robot just takes care of you. Will we achieve that?
I think, at some point, absolutely. I think that’s going to happen. Whether it will happen in the next 10 years, that’s something that’s probably debatable. The ethics involved in these things are complicated. It’s not as simple. Well, the data says it’s– like the article, where hey, it does better than the humans, but still, the humans have something that AI doesn’t, and that’s the intuition piece.
And actually, it was really interesting. I attended a presentation yesterday. They were talking about built-in bias into artificial intelligence. Yeah, we’re all biased. And when we write code, it’s probably going to have bias. So we have a long way to go, but it is encouraging that we have tools that are going to help distill data down to a point to make it so that the subject matter experts can really concentrate on the subject, and not on sifting through massive volumes of data.
MARK SCHAEFER: Well, I don’t think we can finish the show without talking about the TGen Foundation and the incredible work that it’s doing, and the fact that it’s ultimately funding the work that’s getting done. How can companies or individuals get involved to support and learn more about the TGen Foundation?
JAMES LOWEY: Absolutely. They are a critical aspect to TGen, and you can actually– if you go to the TGen website at www.tgen.org, there are links on the website to the Foundation where you can connect with some of those folks. So hey, we welcome everybody to come help. I mean, at the end of the day, it’s going to benefit all of us. So in my opinion, it’s a no-lose value proposition.
MARK SCHAEFER: Well, that’s amazing, James. Thank you so much for your work, for your dedication, and for the really inspiring discussion today. And thanks to all of you for listening to Luminaries , where we talk to the brightest minds in tech. On behalf of Doug Karr, this is Mark Schaefer saying so long for now, and we’ll see you next time on Luminaries .
NARRATOR: Luminaries– talking to the brightest minds in tech, a podcast series from Dell Technologies.