Science in Parallel: A Computational Science Podcast
Science in Parallel focuses on the people of computational science and their interdisciplinary research to simulate climate change and the cosmos, understand viral infections, build alternative energy strategies and more – all using high-performance computing (HPC). We've been shortlisted twice for the Publisher Podcast Awards: for 2022 Best Technology Podcastand for 2023 Best Science and Medical Podcast.
Season 1 celebrated the 30th anniversary of the Department of Energy Computational Science Graduate Fellowship (DOE CSGF) program. Season 2 highlighted how the COVID-19 pandemic has reshaped computational science workplaces and careers, with in-depth conversations about HPC’s future. In Season 3, we explored computing's frontiers: novel processors, tackling big research questions and the practical challenges that come with big data. Season 4 looks at the synergy between creativity and computing. Our guests discuss their creative approaches to climate modeling, quantum physics, neuroscience and materials science and how that mindset shapes their lives outside research, too.
Artificial intelligence is reshaping research to discover new materials for a range of important applications. In this episode, meet Anubhav Jainof Lawrence Berkeley National Laboratory, a researcher who has been at the forefront of this transition. He uses machine learning and other computational tools as a materials scientist to discover compounds that could store and convert energy and solve other societal problems.
Anubhav’s current research path started in graduate school at MIT, where he was supported by a Department of Energy Computational Science Graduate Fellowship. We discuss how computational tools including AI have moved from a novel idea to a central piece of materials discovery, how he applies machine learning tools to other tasks such as mining data from scientific papers, and the rewards that came from writing his blog called Hacking Materials.
This episode concludes our season 4 series on creativity in computing.
From the episode:
As we discussed how the use of machine learning has expanded among materials scientists, Anubhav mentioned the increase in interest at the Materials Research Societyover the last decade. The society shared a five-part series of tutorialson machine learning in materials science presented at recent meetings.
Within the episode we discussed Anubhav’s work on the Materials Projectsince its inception. He also described applications of machine learning for mining materials science data from papers, including the work of former postdoc Vahe Tshitoyan, who is now at Google. He described a project using machine learning for visualizationwith former postdoc Matt Horton, who is now at Microsoft. He also described A-Lab, a project led by Gerd Ceder’s group that uses autonomous robots to synthesize new materials more quickly for experimental testing after they’ve been modeled in computers.
Sometimes extraordinary circumstances like the pandemic offer researchers unexpected opportunities to serve others. Danilo Pérez, now a Ph.D. student in computational neuroscience at New York University, found himself in this situation in Puerto Rico in 2020. He contributed his mathematical modeling expertise as part of a team that built and maintained Puerto Rico’s public health data during that intense period. Later he contributed to AI-based modeling of coronavirus variants that won major honors in the computing community: the 2022 Gordon Bell Special Prize for HPC-Based COVID-19 Research.
These days Danilo is developing computational tools to understand value-based decision making at NYU, a process that can be applied in economics, medicine and public policy. We discuss how compelling science problems have propelled his training, how music and family support him, and his focus on citizen-facing science, especially in Puerto Rico.
Danilo Pérez, a Ph.D. student in computational neuroscientist jointly advised by Christine Constantinopleand Cristina Savinin NYU’s Center for Neural Science. He is a current recipient of a Department of Energy Computational Science Graduate Fellowship (DOE CSGF). This conversation was recorded in July 2023 at the Annual Program Review of the DOE CSGF in Washington, D.C. Read more about Danilo and his work in DEIXIS.
From the episode:
Danilo mentions his high school experiences at CROEM, the Residential Center of Educative Opportunities of Mayagüez. He also described the RISE program, now known as U-RISE, at the University of Puerto Rico-Cayey (UPR-Cayey). That program led to the opportunity to study traumatic brain injury at the University of Missouri-Columbia with Zezong Gu.
He mentioned the importance of Santiago Ramón y Cajal’sclassic drawings of neurons for the field of neuroscience and how that informed the need for artists to help visualize COVID-19 data in Puerto Rico during the pandemic. Danilo became part of the multidisciplinary team that worked on public health data in Puerto Rico during the pandemic, known as COSACO-PR.
For his CSGF practicum in 2022, he worked with an Argonne National Laboratory team led by Arvind Ramanathanon an artificial intelligence model that simulated the evolution of novel coronavirus variants. That team was awarded last year's Gordon Bell Special Prize for High-Performance Computing-Based COVID-19 Research. Read more about the details of that work in this paper.
Danilo talked about the music he makes with his uncle, Yendi Espinal or Yei Flow, in their home studio in New York. You can follow them on Instagram: @lito.trinida and @yeiflowdr.
Portrait credit: Hollenshead/NYU Photo Bureau
Traditional career advice, inside and outside science, often urges people to specialize and become the best at one activity. But that perspective can undervalue interdisciplinary researchers and other polymaths who can see connections between and beyond science and engineering fields. This episode’s guest, Casey Berger, describes how she has navigated this second approach, embracing her many interests, such as science, computing, teaching and storytelling, to make her mark as a physicist and data scientist and as a fiction author.
In the second episode of our podcast series on creativity in computing, Casey talks about her path to physics and computing via Hollywood. She describes the challenges and opportunities of interdisciplinary work, how she pursues her many interests and her advice for building a sustainable, joyful life and career.
Casey Bergeris an assistant professor of physics and data science at Smith Collegein Northampton, Massachusetts. She completed her Ph.D. at the University of North Carolina at Chapel Hill in 2020 and was supported by a Department of Energy Computational Science Graduate Fellowship (DOE CSGF). She earned bachelor’s degrees in physics from Ohio State University and in philosophy and film production from Boston University.
Casey is also a science fiction author. Her latest novel Sister from the Multiverse, part of the Choose Your Own Adventure series, was published in October 2023. This conversation was recorded in July 2023 at the Annual Program Review of the DOE CSGF in Washington, D.C.
Season 4 of Science in Parallel centers around creativity and computing, starting with an interview about climate modeling.
At this nexus of physics, earth science, mathematics and computing, researchers are also racing against the clock—to accurately predict how global climate is shifting before the changes happen. Pulling all the scientific pieces together and communicating those results so that others can use them are significant creative challenges—ones that both Tapio Schneider and Emily de Jong of California Institute of Technology have embraced.
In our conversation, Tapio and Emily describe how both the science and societal impact of climate modeling motivate them, how outdoor activities and music shape their perspectives, and how they view creativity both inside and outside the lab. Later in the episode, Tapio shares his experience as a science advisor to the ClimateMusic Project—an artists’ collaboration that’s producing music and video pieces that explore climate change and solutions to the climate crisis.
Tapio Schneideris a professor of environmental science and engineering at Caltech. He’s a member of the Climate Modeling Alliance (CLiMA)a team of scientists, engineers and applied mathematicians from Caltech, MIT and NASA’s Jet Propulsion Laboratory working on a new earth system model that uses computational and data-science tools to harness Earth observations and make more accurate climate predictions. He spoke about that research at the July 2023 Department of Energy Computational Science Graduate Fellowship (DOE CSGF) Annual Program Review.
Emily de Jongis a Ph.D. student in mechanical engineering at Caltech working in Tapio’s research group. She is a DOE CSGF recipient, who completed her undergraduate degree at Princeton University in 2019.
From the episode:
Tapio serves as a science advisor for the ClimateMusic project, an artists’ collaboration that’s producing music and video pieces that explore climate change and solutions to the climate crisis. This episode includes clips from music that Tapio consulted on.
The first clip used in the episode was from a violin concerto by Theodore Wiprud.
Tapio mentions the influential science essay, More Is Different, by Philip W. Anderson, which was published in Science in 1972 (subscription required). Anderson shared the 1977 Nobel Prize in Physics.
He also mentions GFDL NOAA climate
modelsthat he used during his Ph.D. research
at Princeton. Tapio briefly discusses some of the figures in the early history of climate modeling, such as John von Neumann, Jule Charneyand Norm Phillips.
Emily describes how her work on cloud microphysics has shaped her perspective of the clouds she sees on cycling trips, as in this photo that she took overlooking the Pacific Ocean, or how weather patterns impact her success when mountaineering. She says, “It almost gives me a greater sense of control to have an understanding of what’s going on and why.”
This episode also includes clips from the song “I Wanna Be Cool”by Will Kimbroughand Brant Miller. The song is a central piece of the new Be Cool! climate action campaignby The ClimateMusic Project and Music Declares Emergency.
Making sense of computational science takes a multidisciplinary team, including science visualization experts who translate data into images that both parse information so that it’s comprehensible and render it into beautiful images and skillful animations. Joe Insley of Argonne Leadership Computing Facility and Northern Illinois University has been doing this work for more than 20 years, leveraging deep training in both digital art and computer science to build showstopping visualizations.
We talked about his training, how he approaches this work and how in situ visualization—techniques that allow computational researchers to sift through data as it’s processed—is changing with ever larger supercomputers.
There's a brief companion article and narrated animationavailable at DEIXIS.
Joe Insley is team lead for visualization and data analysis at Argonne Leadership Computing Facilityand associate professor in the School of Art and Design at Northern Illinois University. Joe got his start in scientific visualization creating interactive data explorations for the CAVE (cave automatic virtual environment).
In our discussion about in situ visualization, Joe mentioned his work on the Early Science Programfor Argonne Leadership Computing Facility’s Aurora supercomputer. The project studying the fluid dynamics of blood flow is in collaboration with Amanda Randles of Duke University. Amanda was a Season 2 guest on this podcast.
Joe mentioned several scientific visualization software packages:
Northern Illinois University students joined Joe’s Argonne team as they visualized simulations of turbulent flow within an internal combustion engine.
Joe described some of his visualization process while discussing a supernova explosion simulation by Adam Burrows and his Princeton University team. You can view that as part of a narrated animationthat we produced to accompany this episode.
Middle image: Blood flow simulation created by Joe Insley in collaboration with Amanda Randles, Duke University.
Even after enjoying her first computer science course, Margaret Lawson wasn’t convinced she’d have a place in the field. But today she works on cloud storage for Google after completing her Ph.D. at the University of Illinois, Urbana-Champaign, where she was supported by a Department of Energy Computational Science Graduate Fellowship (DOE CSGF).
This conversation was recorded at the Supercomputing meeting (SC22) in Dallas in November 2022, where Margaret co-led a Birds of a Feather (BoF) session on Ethics in High-Performance Computing. We talked about that session, her pursuit of challenging computer science problems and progress for women in computing.
Margaret Lawson is a software engineer based in Google’s Kirkland, Washington, office. There she primarily works on cloud storage platforms.
During her DOE CSGF practicum, Margaret worked on Ascent, an in situ analysis and visualization library for HPC developed at Lawrence Livermore National Laboratory.
We also talked about her work at Google. At SC22, she did a presentation about how DAOS, an open-source object storage platform developed by Intel, performs on Google’s cloud platform.
We talked about the Ethics in HPC events that Margaret has co-organized at three of the last four Supercomputing meetings with Jakob Lüttgau of the German Climate Computing Center and the University of Hamburg and Jay Lofstead. Elaine Raybourn of Sandia National Laboratories, a Season 2 guest on this podcast, also participated in early discussions.
Portrait credit: Malcolm Smith.
In early December 2022, Lawrence Livermore National Laboratory announced that the National Ignition Facility (NIF) had achieved fusion ignition—a reaction of merging hydrogen isotopes that produced more energy than the lasers put in. High-performance computing is an important part of designing, analyzing and refining these experiments, and this episode examines the connection between computing and fusion energy.
Tammy’s scientific expertise is doing experiments rather than simulations, but in her current role she considers all parts of the fusion puzzle. She’s at the forefront of one of science and society’s grand challenges: Can we produce clean, sustainable fusion energy on the scale needed to power our planet? Tammy talks about computing’s role in understanding and optimizing fusion reactions and how computing’s crossroads could shape fusion’s future.
From this episode:
U.S. Secretary of Energy Jennifer Granholm’s comments are from the Department of Energy press conference held on December 5, 2022.
We referenced the most recent steppingstone toward inertial fusion ignition that occurred in August 2021. In that near-ignition experiment, the fusion reaction produced 70% of the laser energy that went into the shot.
Tammy also was interviewed as part of a 60 Minutes segment that aired in January 2023 about the fusion ignition announcement and fusion energy research.
The fusion ignition achievement was widely covered across science media outlets and in mainstream media. Here’s a small sample of those articles:
- The New York Times: “Scientists Achieve Nuclear Fusion Breakthrough with a Blast of 192 Lasers”
- The American Institute of Physics: “National Ignition Facility Achieves Long-Sought Fusion Goal”
- Nature: “Nuclear-fusion lab achieves ‘ignition’: what does it mean?”
- Science: “With historic explosion, a long sought fusion breakthrough”
Although he’s always loved space, Gabriel Casabona pursued other fields, including medicine and religion, before landing in astrophysics. We discussed how his passion for big questions motivated him to deepen his knowledge of math and computing, how gravity’s mysteries define his work and other big challenges he hopes to work on during his career.
Gabriel Casabonais a Ph.D. student in computational and theoretical astrophysics at Northwestern University. His work is supported by a Department of Energy Computational Science Graduate Fellowship. This conversation was recorded in person in November 2022 at the SC22 meeting in Dallas, Texas.
From this episode:
Gabriel mentioned Brian Greeneas an influential figure in his path toward theoretical astrophysics. Greene is the director of Columbia University’s Center for Theoretical Physics, a bestselling author and host of two NOVA miniseries.
The APS Bridge Programis an American Physical Societyinitiative that supports students from underrepresented communities. The program lasts up to two years and supports post-baccalaureate students with research experience, advanced coursework and mentoring as they apply to doctoral programs in physics.
As he was talking about the death of stars, Gabriel briefly referred to Thor’s hammer, which in Marvel Comics was forged from a metal from the heart of a dying star.
Gabriel mentioned GW170817— an astrophysical event detected in August 2017. Two neutron stars collided — the first cosmic event that astronomers observed via electromagnetic radiation, neutrinos and gravitational waves. The Laser Interferometer Gravitational-Wave Observatory(LIGO) observes gravitational waves from the Earth’s surface. Gabriel is a member of the Laser Interferometer Space Antenna(LISA) mission — a partnership between the European Space Agency and NASA — that will measure gravitational waves using three orbiting spacecraft.
The exascale era in computing has arrived, and that brings up the question of what’s next. We’ll discuss a few emerging processor technologies — molecular storage and computing, quantum computing and neuromorphic chips — with an expert from each of those fields. Learn more about these technologies’ strengths and challenges and how they might be incorporated into tomorrow’s systems.
- Luis Ceze, professor of computer science at the university of Washingtonand CEO of the AI startup OctoML.
- Bert de Jong, senior scientist and department head for computational sciences at Lawrence Berkeley National Laboratoryand deputy director of the Quantum Systems Accelerator. Bert entered computing from his research in theoretical chemistry.
- Catherine (Katie) Schuman, a neuromorphic computing researcher and an assistant professor of computer science at the University of Tennessee, Knoxville.
From the episode:
Luis Ceze was an organizer and co-author of this 2016 report: Arch2030: A Vision of Computer Architecture Research over the Next 15 Yearsby the Computing Community Consortium, supported by funding from the National Science Foundation.
Sarah Webb’s conversation with Bert de Jong was also the basis for an ASCR Discovery article, “Quantum Evolution,” that was published to coincide with World Quantum Day on April 14. The Quantum Systems Accelerator published this impact reportin March.
You can read more about how quantum computing could help materials science and high-energy physics in these papers by Bert de Jong and his colleagues: “Simulating Quantum Materials with Digital Computers” and “Quantum Simulation for High-Energy Physics.”
Katie Schuman was the lead and corresponding author of a 2022 perspective article in Nature Computational Science: “Opportunities for neuromorphic computing algorithms and applications.”
For more information about the Ceze group’s molecular computing research, check out these Nature Communications papers: “Probing the physical limits of reliable DNA data retrieval”and “Molecular-level similarity search brings computing to DNA data storage.” Melissa Queen, a 2018-2022 DOE CSGF recipient, is a co-author of the second paper.
In our unedited conversation, Luis specifically mentioned that groups from the University of Illinois, Urbana-Champaign, Harvard University, Cambridge University and ETH Zurich are studying molecular storage. He also mentioned the DNA Data Storage Alliance.
In our discussion of potential milestones for quantum computing, Bert mentioned simulating the FeMoco systemfor reducing nitrogen to ammonia for fertilizer. FeMocois an abbreviation for iron-molybdenum co-factor. This molecule is an important part of biological nitrogen fixation catalyzed by nitrogenase enzymes, the only known biological systems capable of converting abundant atmospheric nitrogen to ammonia. Today, most ammonia for fertilizer is produced via the Haber-Bosch process, a chemical cascade that consumes approximately 1% of the world’s total energy production and significantly contributes to global carbon-dioxide emissions.
Bert de Jong headshot credit: Berkeley Lab/Quantum Systems Accelerator
In our first two episodes of Science in Parallel’s Season 2, we’ll be talking about how the pandemic pivot to remote work marks a turning point in workplace structure for many computational scientists. We talk with computational scientists who worked remotely about what they struggled with, what functioned well and the lessons they’ll take into the future.
In this first part, we’ll also focus on the social science of how people experienced remote work.
Jerry Wang is an assistant professor of civil and environmental engineering at Carnegie Mellon University. He was a DOE CSGF recipient from 2014 to 2018 while pursuing his Ph.D. at Massachusetts Institute of Technology. Jerry works on particle-based simulations to study soft and active matter, for applications ranging from nanoscale devices to pedestrian mobility.
Elaine Raybourn is a social scientist in Sandia National Laboratories’ Applied Information Sciences Center. She is also an institutional principal investigator for one of the DOE Exascale Computing Project’s many individual research teams: Sandia’s interoperable design of extreme-scale application software (IDEAS) team. IDEAS focuses on team of teams, software developer productivity and software sustainability.
From the episode:
- Elaine has organized the ECP’s Strategies for Working Remotely panel series since 2020. Check out their slides and videos about topics such as setting up a home office space, parenting, working with interns and hybrid work.
- The increased use of video conferencing during pandemic lockdowns highlighted the problem of degraded communication, a concept that is commonly called “Zoom fatigue.”
- You can also read more from Elaine about how ECP members experienced remote work and how they coped with the loss of office whiteboards.
- A version of the interview with Elaine Raybourn is available as an ASCR Discovery article.
In Season Two of Science in Parallel, we’re examining how pandemic shutdowns have reshaped computational science workplaces. In our last episode we focused on the effects of virtual work and how the Exascale Computing Project’s Strategies for Working Remotely panel series fostered communication and creativity. This episode brings in additional stories from graduate students, a professor and an early career researcher at a DOE national lab about the challenges and benefits of remote work.
Jason Torchinsky is a Ph.D. student in applied mathematics at the University of Wisconsin-Madison and a third-year DOE CSGF recipient. They work on methods for applying parallel computing in climate models, particularly integrating disparate models to simulate the Madden-Julian Oscillation, an area of high and low moisture that moves around the Earth’s atmosphere every 30 to 60 days.
Hilary Egan joined the National Renewable Energy Laboratory’s Computational Science Center as a data scientist in June 2020. Hilary completed her Ph.D. in astrophysics and planetary science at the University of Colorado Boulder and was a DOE CSGF recipient from 2014 to 2018. Hilary works on AI for scientific computing across applications including materials science, data center efficiency, and building retrofits.
Laura Nichols is a second-year DOE CSGF recipient and a Ph.D. student in computational solid-state physics at Vanderbilt University. She uses quantum mechanics to model how defects in semiconductor devices are activated and lead to degradation. Laura is incorporating that model into her group’s code that describes defect-related processes such as scattering and electron capture.
Pandemic work was especially challenging for computational scientist parents, who often juggled new work arrangements while balancing their children’s care and education. In this episode you’ll hear from a couple who were Ph.D. students and had a 10-month-old baby when lockdowns sent them all home in March 2020. The situation challenged their work and their mental health. As they adapted to these experiences, they changed career paths and their perspectives on life and work.
Kalin Kiesling is a nuclear engineer in the nuclear science and engineering division at Argonne National Laboratory. Her work focuses on the development of computational tools used to design the next generation of nuclear reactors. Prior to joining Argonne, Kalin earned her Ph.D., M.S., and B.S. in nuclear engineering and engineering physics from the University of Wisconsin-Madison.
Brian Cornille is a member of technical staff at Advanced Micro Devices. He works on porting and performance optimization of scientific applications targeting AMD platforms, such as Frontier at Oak Ridge National Laboratory and the upcoming El Capitan at Lawrence Livermore National Laboratory. Brian was a DOE CSGF recipient from 2016 to 2020 and completed both a B.S. and Ph.D. in nuclear engineering and engineering physics at the University of Wisconsin-Madison.
Content warning: This episode discusses mental health and miscarriage.
After COVID-19 lockdowns and 2020 wildfires near his Oregon home, computational scientist Jeff Hammond decided to make big moves. In 2021, his family of five emigrated from Portland to Finland, and Jeff changed positions, leaving Intel and taking a new job with NVIDIA. Even before 2020, he had worked primarily remotely and discusses the lessons he hopes technology companies learn from pandemic work.
Jeff Hammond, a principal engineer with NVIDIA, is affiliated with the company’s office in Helsinki, Finland. From 2014 to 2021, Jeff worked for Intel, and was based in Portland, Oregon. Prior to that he worked at Argonne National Laboratory. Jeff was a Department of Energy Computational Science Graduate Fellowship recipient from 2005 to 2009 at the University of Chicago and focused on developing open-source chemistry simulation software, NWChem, with Karol Kowalski at Pacific Northwest National Laboratory.
- In the episode, host Sarah Webb mentions a study that talks about commuting hours saved in the United States during pandemic lockdowns.
- Curious about learning Finnish? Read more about Jeff’s experiences over the last year.
Valerie Taylor doesn’t shy away from challenging problems. At Argonne National Laboratory, she manages teams that develop algorithms, data management strategies, software and hardware to support scientific simulations, including those on the Department of Energy’s leadership-class supercomputers. Her research focuses on performance analysis—the factors involved in making computations efficient. On top of that, she maintains a parallel line of work supporting computer scientists from historically marginalized communities toward building a more diverse computing workforce.
You’ll hear Valerie talk about her career path, what excites her about computing, and the sustained commitment needed to boost diversity, equity and inclusion in this field.
Valerie Taylor is the director of the mathematics and computer science division at Argonne National Laboratory. She moved to Argonne in 2017 after more than 25 years in academia at both Northwestern University and at Texas A&M University. She also is the president and chief executive officer of the Center for Minorities and People with Disabilities in IT (CMD-IT), a non-profit dedicated to supporting historically marginalized communities in computing. She has been recognized with numerous awards, both for her research and her work advocating for underrepresented communities in computing.
From the episode:
- Valerie is the principal investigator for Threadwork, a DOE-funded microelectronics project that brings together materials scientists, computer scientists, software developers and physicists to interact across disciplines to design chips, devices and detectors for applications such as high energy physics.
- In the discussion about her Ph.D., Valerie described the community among Black Ph.D. students during her years at the University of California, Berkeley and how that support has continued through their careers. She noted how Sheila Humphreys supported their work. In our unedited conversation she mentioned Colin Parris, a senior vice-president and chief technology officer at General Electric; Gary May, chancellor at the University of California, Davis; and Kevin Kornegay and Arlene Cole, both professors at Morgan State University.
- Both CMD-IT’s Tapia Conference and Anita B.Org’s Grace Hopper Celebration are held annually in September.
- Sarah asked Valerie about this 2018 article that she cowrote in IEEE Computer: “Increasing Women and Underrepresented Minorities in Computing: The Landscape and What You Can Do.”
- Valerie has cochaired the Committee on Women in Science, Engineering and Medicine at the National Academies of Sciences, Engineering and Medicine, which produced this recent report: “Transforming Trajectories for Women of Color in Tech.”
- An article based on this interview is available at ASCR Discovery.
Science in Parallel’s second season concludes with a conversation about answering important questions in biology and medicine with leadership class supercomputers, including urgent issues that came up during the COVID-19 pandemic. You’ll hear from Anda Trifan of the University of Illinois at Urbana-Champaign (UIUC) and Amanda Randles of Duke University.
Both Anda and Amanda were awarded a Department of Energy Computational Science Graduate Fellowship to pursue doctoral studies. Between them, they have worked on a total of five projects that have been finalists for either the Association of Computing Machinery (ACM) Gordon Bell Prize or the Special Prize for COVID-19 research. Adding to the excitement of their pandemic work: They both navigated the at-home adventure of raising very young children during lockdown. They talk about what drives them, the challenge of working at the cutting edge of HPC and biology and medicine, and their advice for other researchers, particularly other women in science.
Anda Trifan is a UIUC graduate student who uses molecular dynamics simulations to study the interactions between proteins and membranes in cancer and other diseases. Through her Argonne National Laboratory practicum, she joined the teams that modeled the SARS-CoV-2 spike protein and virus aerosols. And Anda was first author on a third paper that simulated part of the SARS-CoV-2 replication process. All three projects were finalists for the ACM Gordon Bell Special Prize for High Performance Computing-Based COVID-19 Research, and the spike protein paper won the 2020 award.
Amanda Randles is an assistant professor in biomedical engineering at Duke University. Much of her work has focused on large-scale personalized blood flow simulations, using medical imaging and data to build a three-dimensional image of an individual’s unique vessel geometry. Such tools one day could help doctors diagnose and treat disease. Some of that research, for their blood flow simulator known as HARVEY, was a finalist for an ACM Gordon Bell Prize in 2015. She has received numerous other honors including the ACM Grace Murray Hopper Award in 2017.
- Anda has worked with UIUC's Emad Tajkhorshid on the k-ras protein for her Ph.D. research. She mentioned the molecular dynamics tools, NAMD and VMD, that were developed at UIUC.
- Arvind Ramanathan of Argonne National Laboratory advised Anda’s DOE CSGF practicums. Anda also mentioned their collaborators on their COVID-19 work: Rommie Amaro’s lab at the University of California San Diego and Lillian Chong’s lab at the University of Pittsburgh.
- For more information on the coronavirus projects, here’s a UCSD press release about their simulations showing that gate-like sugars on the surface of the SARS-CoV-2 spike protein are critical for infection. The New York Times covered the work on simulating COVID-19 aerosol droplets.
- Anda was featured in a profile article in the 2022 issue of DEIXIS magazine.
- Amanda discussed how her experiences working on the IBM Blue Gene system helped to shape her interests. Her work as a Lawrence Fellow at Lawrence Livermore National Laboratory laid the foundation for her work at Duke.
- In the early days of the pandemic, the COVID-19 HPC Consortium brought together industry, academia, federal agencies, DOE national labs and international partners to accelerate pandemic research. That effort funded Amanda’s ventilator project and part of Arvind Ramanathan’s work at Argonne among its 114 projects.
- Amanda mentioned this Microsoft story about their ventilator project. This ASCR Discovery article describes her team’s work on modeling circulating tumor cells using HARVEY.
- In her discussion about the challenges of working at the cutting edge of both HPC and biomedicine, Amanda mentioned the difficulty in modeling blood flow in the human body beyond a single heartbeat. In October, after the episode was recorded, she received a prestigious 2022 National Institutes of Health Director’s Pioneer Award to pursue this question.
Jeff Hittinger of Lawrence Livermore National Laboratory embodies the term scientist-chimera. He talks about the many scientific hats he’s worn simultaneously – computer scientist, applied mathematician and physicist. As director for the Center for Applied Computing (CASC) and as co-principal investigator for the DOE CSGF, he wears many more. He talks about scientific success, leadership and the tricks he’s cultivated for communicating science to broader audiences through the Livermore Ambassador Lecture series.
Jeff was a DOE CSGF recipient from 1996 to 2000 while earning his Ph.D. in aerospace engineering and scientific computing at the University of Michigan. He was one of the first recipients of the Frederick A. Howes Scholar Award and received the 2021 James Corones Award in Leadership, Community Building and Communication.
One of today’s hottest areas of computational research could help build better solutions for one of global society’s steepest challenges. Three early career computational scientists – Priya Donti, Kelly Kochanski and Ben Toms – talk about AI’s potential for understanding and predicting climate shifts, supporting strategies for incorporating renewable energy, and engineering other approaches that reduce carbon emissions. They also describe how AI can be misused or can perpetuate existing biases.
Working at this important research interface requires broad knowledge in areas such as climate science, public policy and engineering coupled with computational science and mathematics expertise. These early career researchers talk about their approaches to bridging this gap and offer their advice on how to become a scientific integrator.
Priya Donti is a Ph.D. student at Carnegie Mellon University, pursuing a dual degree in public policy and computer science, and a fourth-year DOE CSGF recipient (at the time of recording). She is also a co-founder and chair of the volunteer organization, Climate Change AI, which provides resources and a community for researchers interested in applying artificial intelligence to climate challenges. Priya was named to MIT Technology Review’s 2021 list of Innovators Under 35. Read more about Priya and her work in the 2021 issue of DEIXIS.
Kelly Kochanski completed a Ph.D. in geological sciences at the University of Colorado, Boulder in 2020 and works as a senior data scientist in climate analytics at McKinsey & Company. Kelly was a DOE CSGF recipient from 2016 to 2020, and her graduate research was featured in the DEIXIS 2020. She also is profiled in the 2021 issue as one of this year’s recipients of the Frederick A. Howes Scholar Award.
Ben Toms also finished his Ph.D. last year at Colorado State University studying atmospheric science and is a 4th year DOE CSGF recipient. He has founded a company, Intersphere, that provides weather and climate forecasts up to a decade into the future.
From the episode:
- Kelly and Priya contributed to the review article: Tackling Climate Change with Machine Learning, which was published on the arXiv preprint server in 2019.
- In the discussion about interpretable AI, Priya mentioned an article by Cynthia Rudin: Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead.
- Ben mentioned Vulcan’s work to build faster climate change models.
Curiosity, mentors and a summer working in concrete with his grandfather shaped Quentarius Moore’s science career studying 2-D materials. He recently completed his fourth year as a DOE CSGF recipient, while pursuing a chemistry Ph.D. at Texas A&M University. He completed both his bachelor's and master's degrees in chemistry at Jackson State University in Mississippi. Read more about Quentarius and his graduate research in the 2021 issue of DEIXIS.
Quentarius was a co-author on this recent ACS Nano paper about molybdenum disulfide structures published by researchers from Texas A&M and Sandia National Laboratories.
From the Episode:
- Quentarius is a member of the NSF Center for the Mechanical Control of Chemistry (CMCC), which is led by James Batteas of Texas A&M. Funded initially by a $1.8 million grant, researchers focus on ways to apply precise forces to coax or alter chemical reactions at surfaces and interfaces. The CMCC includes researchers from Texas A&M; the University of California, Merced; CUNY Advanced Science Research Center; Northwestern University; the University of Pennsylvania; and McGill University.
- He also mentioned the Louis Stokes Alliances for Minority Participation Bridge to Doctorate program, conferences hosted by the National Organization for the Professional Advancement of Black Chemists and Chemical Engineers (NOBCChE) and the beautiful crystals of his favorite element, bismuth.
Middle image: Quentarius tells how his grandfather’s patience and persistence guided his work ethic, including a story about this 20-by-20-foot plot that became a productive garden in summer 2015.
Bottom image: Quentarius built this computer workstation with his younger brother in 2020 during the COVID-19 pandemic.
Alicia Magann got her start in control systems engineering research, exploring tools for directing large-scale chemical processes. As a Ph.D. student, she turned the dials of quantum chemistry in Herschel Rabitz’s research group at Princeton University with support from the DOE CSGF. She talks about her work on quantum algorithms, her cross-country road trip from New Jersey to her practicum in California and how her dad is her scientific hero.
Read more about Alicia and her work in the 2021 issue of DEIXIS.
Alicia recently co-authored a research paper about how some of these quantum control principles could be used to inspire advances in quantum computing. The perspective paper was published in the journal PRX Quantum.
Avoiding the changing climate’s most extreme impacts will require a technological revolution to power daily life from renewable sources. An entrepreneur, an engineering professor and a DOE-laboratory materials scientist – all DOE CSGF and Massachusetts Institute of Technology alumni – discuss technical challenges from nuclear energy to heat transfer to hydrogen generation and the importance of choosing high-impact research problems. In addition to talking about science, engineering and computation, they highlight the need for a strong social and political movement to drive a complete overhaul of our energy infrastructure.
Leslie Dewan is a nuclear engineering entrepreneur and venture capitalist, who is currently the CEO of RadiantNano, a startup focused on radiation detection, identification and imaging. When this conversation was recorded, she headed a venture capital fund, Criticality Capital. She co-founded the alternative nuclear reactor startup Transatomic Power in 2010 and served as its CEO for eight years. Leslie was a DOE CSGF recipient from 2010 to 2013.
Asegun Henry is an MIT associate professor of mechanical engineering. Besides his academic posts, he completed postdoctoral research in materials theory at Oak Ridge National Laboratory and was a fellow in the Advanced Research Projects Agency- Energy (ARPA-E). His research group has contributed to the fields of solar fuels, thermochemistry and phonon transport and has developed an all-ceramic, ultra-high-temperature mechanical pump, which could support renewable energy innovations including grid-level storage. Asegun was a DOE CSGF recipient from 2005 to 2009.
Brandon Wood is the associate program lead for Hydrogen and Computational Energy Materials at Lawrence Livermore National Laboratory. He is also deputy director of the Laboratory for Energy Applications for the Future (LEAF). Brandon works on simulation techniques for studying energy storage and conversion, particularly related to hydrogen technologies such as water-splitting catalysts and solid-state batteries. He also models processes such as corrosion that affect energy system performance. Brandon was a DOE CSGF recipient from 2003 to 2007.
Additional reading and episode notes:
- Leslie mentioned ARPA-E’s program to develop digital twins of advanced nuclear reactors. In part of our conversation that isn’t included here, she mentioned Vannevar Bush’s 1967 book, Science is Not Enough, which drives home the importance of communication and telling people the underlying why behind the science.
- Asegun was the lead author on this 2020 Nature Energy commentary: Five thermal energy grand challenges for decarbonization. MIT News published this profile in July 2021.
- Brandon and his colleagues recently published this paper about improved computational tools for modeling metal hydrides for hydrogen storage. He was also a coauthor on this review covering experimental and modeling strategies for studying solid-state hydrogen storage.
Aurora Pribram-Jones works on hot, dense electrons – simulating extreme chemistry that can happen within giant planets like Jupiter or nuclear fusion experiments. Aurora’s career included many initial detours on the way to science, but the flexibility of community college classes and a job at a technical bookstore paved their path toward research. Now a member of the chemistry faculty at the University of California, Merced, Aurora finds purpose in teaching and mentoring students and supporting the whole scientist, especially those from underrepresented and marginalized communities.
Aurora completed a Ph.D. at the University of California, Irvine, and was a DOE CSGF recipient from 2011 to 2015. They carried out postdoctoral research at the University of California, Berkeley, and at Lawrence Livermore National Laboratory, the latter supported by a Lawrence Postdoctoral Fellowship. Aurora received the Frederick A. Howes Scholar Award in Computational Science in 2016.
- Much of Aurora’s research uses density functional theory (DFT), a method for modeling molecules, atoms and subatomic particles that incorporates quantum mechanics. Aurora co-authored this review article that provides a DFT history and overview. One ongoing project in their group is focused on developing nontoxic, lead-free copper alloys for plumbing, a collaboration that includes researchers at Harvey Mudd College and the University of New South Wales as well as DOE CSGF alumnus Jonas Kaufman. A recent paper on that research was published in Physical Review Materials earlier this year.
- Aurora discussed the backstory for their research group’s website, hypugaea.com, and why they chose the scientific name of a burrowing owl as a space for a group that focuses on physical science. Aurora writes: “I am committed to creating spaces that support students and colleagues who do not fit into narrow definitions of who we are and what we care about. You and your lives matter, above- and below-ground.”