Science & Technology

Q&A with Fred Hutch’s new chief data officer on building a data ecosystem at an institutional scale

Incoming Fred Hutch chief data officer Jeffrey Leek. (Photo courtesy of Leek)

Seattle’s Fred Hutchinson Cancer Research Center has recruited a new vice president and chief data officer: biostatistician and data science educator Jeffrey Leek. Leek will start in the newly formed position by July 1, the Hutch announced Tuesday.

Leek will be leaving his current post as a professor of biostatistics and oncology at the Johns Hopkins Bloomberg School of Public Health and co-director of its Data Science Lab.

He develops tools to help understand massive-scale genomic and biomedical data and has helped generate online learning programs that have enrolled more than 8 million people. He also directs Data Trail, which trains and supports underserved communities for work in data science, and in 2020 co-founded data management startup Streamline Data Science.

Leek will arrive at the Hutch as the institution works out a new proposed partnership structure with UW Medicine, Seattle Children’s and the Seattle Cancer Care Alliance, and he will shape an integrated data enterprise. His move is also in some ways a homecoming — the Idaho native earned his Ph.D. at the University of Washington and has kept relationships with regional researchers.

We spoke with Leek about his plans at Fred Hutch, how he sees the institution integrating into the tech and entrepreneurship community, and his plans to expand data science education. The following interview has been edited for brevity and clarity.

GeekWire: Congrats on the new gig, Jeffrey. What are you most excited about?

Jeffrey Leek: I am excited about the chance of building a data ecosystem at an institutional scale. I’ve been leading the Johns Hopkins Data Science Lab here for the last few years, and we do a lot of infrastructure development and data management training, massive online open courses on Coursera. This opportunity is to take that thinking and scale it to the level of an institution and figure out a way to help everybody at the Hutch to use data in ways that are more efficient, easier and fun. The researchers there are total world class, and it’s fun to imagine the new collaborations that we’re going to build. Being a part of the greater Seattle data and tech ecosystem is also just incredibly compelling.

GeekWire: What goes into making a data ecosystem?

Leek: It’s about setting up the right technologies, making it easy to use data, building in the right ethical and privacy concerns, and then training people to do use data in ways that maybe they haven’t been able to before. And so, all of that together creates an environment where you can accelerate a whole bunch of different kinds of research, from cancer research to infectious diseases to everything across the board.

GeekWire: What are the opportunities in bringing together data across Seattle-area research institutions?

Leek: The data that’s being collected about all of the different treatments and protocols at these institutions can be brought together in a way that allows us to do large scale analyses, to identify which therapies work within which sub-populations and identify hypotheses for following up with future research. The real excitement around it is being able to leverage data from those institutions to answer some of the questions that are really hard to answer at the single institution scale. When you have a rare event or side effect, it’s harder to see that.

(Fred Hutch Photo)

GeekWire: How do you see Fred Hutch integrating into the larger tech and entrepreneurship community?

Leek: The Hutch is really thinking through where the points are for a mutually beneficial relationship between the institution and private partners — whether that’s investors to help support spinning out ideas focused on data science, or talking to potential partners about expanding the size and scope of the analyses that we can do. You want to be very careful to maintain the trust of the nonprofit enterprise, while at the same time finding ways to get the best of the private infrastructure to bring to bear on collaborating to build better solutions for patients. It’s a tricky thing to navigate.

GeekWire: How do you see the Hutch working with larger players like Amazon and Microsoft, vs. smaller startups?

Leek: There are differences and opportunities there for sure. With the larger companies, my experience has been, and what I’m looking forward to doing, is starting with small points of contact. So, you find one researcher who has one really serious need, such as scaling a dashboard with infectious disease data to a large community, or maybe you have a large-scale tech partner that already has existing technologies. And then once you’ve had that first relationship built, then you can kind of scale out from there to identify other places you can grow. I found that the thing that really like lights a fire is identifying people who are psyched about a common problem and getting them to work together and build a solution for it. And then pouring the institutional support behind it can really explode cool ideas out into the public.

GeekWire: Can you expand on the Hutch’s role in partnering with or fostering startups?

Leek: There’s everything from licensing technologies to identifying ways to support trainees who see themselves growing into a more startup-like community rather than going into academics. How do we create that support infrastructure, so folks with that kind of passion can be able to do it? It’s all about building relationships between people who want to solve similar problems. I think there’s a lot of room for creativity, especially within the data, machine learning and AI space at the interface between biotechnology and data science.

GeekWire: Tell us more about your outreach efforts with Data Trail and how you feel that will fit into your work in Seattle.

Leek: People who take massive online open courses are already well educated, like master’s level people trying to change careers. And so we started to conceive how could we build a program that would bring people from the community into the data enterprise at Hopkins. We built a training program designed for basic data science and data management and we partnered with community-based nonprofits to identify talented math and science students in our neighborhood. We put them through a tutoring program where we would pay them to complete the courses, provided them with laptops, and worked with nonprofit partners to provide social and financial support. We worked with Johns Hopkins so that they could come in as entry level data science and data managers, and created this pathway into the Johns Hopkins ecosystem. I’m really excited to have Data Trail in Seattle start to grow. We’re hoping to run it in other places around the country as well.

GeekWire: Final thoughts?

Leek: One of the things I’d really love to do in terms of the tech ecosystem is to start building connections to people who are passionate about cancer research and open source data science training, any of the things I’ve talked about in the ecosystem. I hope to be an active participant in that system and the people there. I’m hoping that folks on that side will reach out to me if they’re excited about working with us, and I plan on reaching out in the other direction, too.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *