Research Computing - MGHPCC Feed

Exploring Phytoplankton Diversity

reporting by Helen Hill

In a new paper,  MIT Senior Research Scientist Stephanie Dutkiewicz and collaborators use computers housed at the MGHPCC to develop theories to explain and predict how phytoplankton are distributed in the ocean.

Ocean microbial communities (phytoplankton) play an important role in the global cycling of elements including climatically significant carbon, sulfur, and nitrogen. Photosynthetic microbes in the surface ocean fix carbon and other elements into organic molecules, fueling food webs that sustain fisheries and most other life in the ocean. Sinking and subducted organic matter is remineralized and respired in the dark, sub-surface ocean maintaining a store of carbon about three times the size of the atmospheric inventory of CO2.

The phytoplankton communities sustaining this global-scale cycling are functionally and genetically extremely diverse, non-uniformly distributed and sparsely sampled; their biogeography reflecting selection according to the relative fitness of myriad combinations of traits that govern interactions with the environment and other organisms. Scientists at MIT are using sophisticated computer models to develop theories to explain and predict phytoplankton biogeography.

Stephanie Dutkiewicz is a Senior Research Scientist in MIT’s Center for Global Change Science (CGCS) and a member of Mick Follow’s marine microbe and microbial community modeling group in the Department of Earth, Atmospheric and Planetary Sciences at MIT. Dutkiewicz’s particular research interests lie at the intersection of the marine ecosystem and the physical and biogeochemical environment. She is especially interested in how the interactions of these components of the earth system will change in a warming world. In recent work, Dutkiewicz’s focus has been on how ocean physics and chemistry control phytoplankton biogeography, and how in turn those organisms affect their environment. To do this she pairs complex numerical models with simple theoretical frameworks, guided by laboratory, field and satellite observations.

“Phytoplankton are an extremely diverse set of microorganisms spanning more than seven orders of magnitude in cell volume and exhibiting an enormous range of shapes, biogeochemical functions, elemental requirements, and survival strategies,” Dutkiewicz explains. “This range in traits plays a key role in regulating the biogeochemistry of the ocean, including the export of organic matter to the deep ocean, a process critical in oceanic carbon sequestration contributing to the modulation of atmospheric CO2 levels and climate. Biodiversity is also important for the stability of ecosystem structure and function, though the exact nature of this relationship is still debated. Studies suggest that diversity loss appears to coincide with a reduction in primary production rates and nutrient utilization efficiency, thereby altering the functioning of ecosystems and the services they provide. Diversity is important, but what factors control diversity remains an elusive problem. Diversity is also important for higher trophic levels, with different types/sizes supporting different foodwebs”

While biodiversity of phytoplankton is important for foodwebs, and marine biogeochemistry, the large-scale patterns of that diversity are not well understood and are often poorly characterized in terms of their relationships with factors such as latitude, temperature, and productivity. In a new study, Dutkiewicz and co-authors from MIT, the Institut de Ciencies del Mar, Spain, the National Oceanography Centre, Southampton, UK, and California State University San Marcos, use ecological theory and a numerical global trait-based ecosystem model to seek a mechanistic understanding of those patterns. The paper, using one of MIT’s supercomputing clusters housed at the MGHPCC, appeared in Biogeosciences this month.

Focusing on three dimensions of trait space (size, biogeochemical function, and thermal tolerance), Dutkiewicz et al’s study suggests that phytoplankton diversity is in fact controlled by disparate combinations of drivers: the supply rate of the limiting resource, the imbalance in different resource supplies relative to competing phytoplankton demands, size-selective grazing, and transport by the moving ocean.

Model diversity measured as annual mean normalized richness in the surface layer. Normalization is by the maximum value for that plot (value noted in the bottom right of each panel). (a) Total richness determined by number of individual phytoplankton types that coexist at any location; (b) size class richness determined by number of coexisting size classes; (c) functional richness determined by number of coexisting biogeochemical functional groups; (d) thermal richness determined by number of coexisting temperature norms – Image courtesy: The researchers.

“Using sensitivity studies, we were able to show that each dimension of diversity is controlled by different drivers,” Dutkiewicz explains. “A model including only one (or two) of the trait dimensions will exhibit different patterns of diversity than one which incorporates another trait dimension.”

Dutkiewicz says, “I believe this is one of my more exciting papers, and really addresses some fundamental components of marine biodiversity, as well as highlighting why we need to be very careful in what we are defining as diversity. Our results indicate that trying to correlate diversity with quantities like temperature, or productivity (as is frequently done) is doomed to fail or worse to give wrong answers. There is no one mechanism that controls biodiversity. Understanding at the “dimensions” level is essential.”

Story image: Adapted from phytoplankton microscope image collected on a HOT cruise – courtesy C. Follett/ CBIOMES

About the Researcher Stephanie Dutkiewicz

Stephanie Dutkiewicz, who holds a PhD from the University of Rhode Island (1997) has been at MIT since 1998. She is lead author on a widely reported 2019 study in Nature Communications indicating that climate change will alter the color of the oceans.

Publication

Stephanie Dutkiewicz, Pedro Cermeno, Oliver Jahn, Michael J. Follows, Anna E. Hickman, Darcy A. A. Taniguchi, and Ben A. Ward (2020), Dimensions of marine phytoplankton diversity [link], Biogeosciences, doi: 10.5194/bg-17-609-2020

 

Brainstorming energy-saving hacks on Satori, MIT’s new supercomputer

Three-day hackathon using Satori, the computing cluster IBM donated to MIT last year housed at the MGHPCC, explores methods for making artificial intelligence faster and more sustainable.

Read this story at MIT News

Mohammad Haft-Javaherian planned to spend an hour at the Green AI Hackathon — just long enough to get acquainted with MIT’s new supercomputer, Satori. Three days later, he walked away with $1,000 for his winning strategy to shrink the carbon footprint of artificial intelligence models trained to detect heart disease.

Top prize-winners in the Green AI Hackathon were (clockwise, from top left): Mohammad Haft-Javaherian, Alex Andonian, Camilo Fosco, and Jonathan Frankle.
Photo panel: Samantha Smiley

“I never thought about the kilowatt-hours I was using,” he says. “But this hackathon gave me a chance to look at my carbon footprint and find ways to trade a small amount of model accuracy for big energy savings.”

Haft-Javaherian was among six teams to earn prizes at a hackathon co-sponsored by the MIT Research Computing Project and MIT-IBM Watson AI Lab Jan. 28-30. The event was meant to familiarize students with Satori, the computing cluster IBM donated to MIT last year, and to inspire new techniques for building energy-efficient AI models that put less planet-warming carbon dioxide into the air.

Chekuri Choudary (left), an infrastructure specialist with IBM Cognitive Systems Lab Services, works with MIT graduate student Ravichandra Addanki.
Photo: Kim Martineau

The event was also a celebration of Satori’s green-computing credentials. With an architecture designed to minimize the transfer of data, among other energy-saving features, Satori recently earned fourth place on the Green500 list of supercomputers. Its location gives it additional credibility: It sits on a remediated brownfield site in Holyoke, Massachusetts, now the Massachusetts Green High Performance Computing Center, which runs largely on low-carbon hydro, wind and nuclear power.

A postdoc at MIT and Harvard Medical School, Haft-Javaherian came to the hackathon to learn more about Satori. He stayed for the challenge of trying to cut the energy intensity of his own work, focused on developing AI methods to screen the coronary arteries for disease. A new imaging method, optical coherence tomography, has given cardiologists a new tool for visualizing defects in the artery walls that can slow the flow of oxygenated blood to the heart. But even the experts can miss subtle patterns that computers excel at detecting.

At the hackathon, Haft-Javaherian ran a test on his model and saw that he could cut its energy use eight-fold by reducing the time Satori’s graphics processors sat idle. He also experimented with adjusting the model’s number of layers and features, trading varying degrees of accuracy for lower energy use.

MIT brain and cognitive sciences graduate students Jenelle Feather and Kelsey Allen were recognized for applying a technique that drastically simplifies models by cutting their number of parameters.
Photo: Kim Martineau

A second team, Alex Andonian and Camilo Fosco, also won $1,000 by showing they could train a classification model nearly 10 times faster by optimizing their code and losing a small bit of accuracy. Graduate students in the Department of Electrical Engineering and Computer Science (EECS), Andonian and Fosco are currently training a classifier to tell legitimate videos from AI-manipulated fakes, to compete in Facebook’s Deepfake Detection Challenge. Facebook launched the contest last fall to crowdsource ideas for stopping the spread of misinformation on its platform ahead of the 2020 presidential election.

If a technical solution to deepfakes is found, it will need to run on millions of machines at once, says Andonian. That makes energy efficiency key. “Every optimization we can find to train and run more efficient models will make a huge difference,” he says.

To speed up the training process, they tried streamlining their code and lowering the resolution of their 100,000-video training set by eliminating some frames. They didn’t expect a solution in three days, but Satori’s size worked in their favor. “We were able to run 10 to 20 experiments at a time, which let us iterate on potential ideas and get results quickly,” says Andonian.

As AI continues to improve at tasks like reading medical scans and interpreting video, models have grown bigger and more calculation-intensive, and thus, energy intensive. By one estimate, training a large language-processing model produces nearly as much carbon dioxide as the cradle-to-grave emissions from five American cars. The footprint of the typical model is modest by comparison, but as AI applications proliferate its environmental impact is growing.

One way to green AI, and tame the exponential growth in demand for training AI, is to build smaller models. That’s the approach that a third hackathon competitor, EECS graduate student Jonathan Frankle, took. Frankle is looking for signals early in the training process that point to subnetworks within the larger, fully-trained network that can do the same job. The idea builds on his award-winning Lottery Ticket Hypothesis paper from last year that found a neural network could perform with 90 percent fewer connections if the right subnetwork was found early in training.

The Green AI Hackathon was organized by IBM’s John Cohn (left) and MIT’s Christopher Hill.
Photo: Kim Martineau

The hackathon competitors were judged by John Cohn, chief scientist at the MIT-IBM Watson AI Lab, Christopher Hill, director of MIT’s Research Computing Project, and Lauren Milechin, a research software engineer at MIT.

The judges recognized four other teams: Department of Earth, Atmospheric and Planetary Sciences (EAPS) graduate students Ali Ramadhan, Suyash Bire, and James Schloss, for adapting the programming language Julia for Satori; MIT Lincoln Laboratory postdoc Andrew Kirby, for adapting code he wrote as a graduate student to Satori using a library designed for easy programming of computing architectures; and Department of Brain and Cognitive Sciences graduate students Jenelle Feather and Kelsey Allen, for applying a technique that drastically simplifies models by cutting their number of parameters.

IBM developers were on hand to answer questions and gather feedback.  “We pushed the system — in a good way,” says Cohn. “In the end, we improved the machine, the documentation, and the tools around it.”

Going forward, Satori will be joined in Holyoke by TX-Gaia, Lincoln Laboratory’s new supercomputer. Together, they will provide feedback on the energy use of their workloads. “We want to raise awareness and encourage users to find innovative ways to green-up all of their computing,” says Hill.

Story image: About two dozen students participated in the Green AI Hackathon, co-sponsored by the MIT Research Computing Project and MIT-IBM Watson AI Lab – Image credit: Kim Martineau

 

Related

IBM gives artificial intelligence computing at MIT a lift MIT News

January Publications

Below is a selection of papers that appeared in December 2019 reporting the results of research using the Massachusetts Green High Performance Computing Center (MGHPCC), or acknowledging the use of Harvard’s Odyssey Cluster, Northeastern’s Discovery Cluster, the Boston University Shared Computing Cluster and MIT’s Engaging Cluster all of which are housed at the MGHPCC.

João Alves, Catherine Zucker, Alyssa A. Goodman, Joshua S. Speagle, Stefan Meingast, Thomas Robitaille, Douglas P. Finkbeiner, Edward F. Schlafly and Gregory M. Green (2020), A Galactic-scale gas wave in the solar neighborhood, Nature, doi: 10.1038/s41586-019-1874-z

Brian Arnold, Mashaal Sohail, Crista Wadsworth, Jukka Corander, William P Hanage, Shamil Sunyaev, Yonatan H Grad (2020), Fine-Scale Haplotype Structure Reveals Strong Signatures of Positive Selection in a Recombining Bacterial Pathogen, Molecular Biology and Evolution, doi: 10.1093/molbev/msz225

Amir Bitran, William M. Jacobs, Xiadi Zhai, and Eugene Shakhnovich (2020), Cotranslational folding allows misfolding-prone proteins to circumvent deep kinetic traps, PNAS, doi: 10.1073/pnas.1913207117

Andrés N Salcedo, Benjamin D Wibking, David H Weinberg, Hao-Yi Wu, Douglas Ferrer, Daniel Eisenstein, Philip Pinto (2020), Cosmology with stacked cluster weak lensing and cluster–galaxy cross-correlations, Monthly Notices of the Royal Astronomical Society, doi: 10.1093/mnras/stz2963

Debjani Sihi, Eric A. Davidson, Kathleen E. Savage, Dong Liang (2020), Simultaneous numerical representation of soil microsite production and consumption of carbon dioxide, methane, and nitrous oxide using probability distribution functions, Global Change Biology, doi: 10.1111/gcb.14855

Lindsay J. Underhill, W. Stuart Dols, Sharon K. Lee, M. Patricia Fabian and Jonathan I. Levy (2020), Quantifying the impact of housing interventions on indoor air quality and energy consumption using coupled simulation models, J Expo Sci Environ Epidemiol., doi: 10.1038/s41370-019-0197-3

V.V. Volkov, R. Chelli, R. Righini, C.C. Perry (2020), Indigo chromophores and pigments: Structure and dynamics, Dyes and Pigments, doi: 10.1016/j.dyepig.2019.107761

Yichao Wang Prof. Sooran Kim Dr. Jingyu Lu Guangyuan Feng Prof. Xin Li (2020), A Study of Cu Doping Effects in P2‐Na0.75Mn0.6Fe0.2(CuxNi0.2‐x)O2 Layered Cathodes for Sodium‐Ion Batteries, Batteries and Supercaps, doi: 10.1002/batt.201900172

Xiao Wu Yi Xu Bradley P. Carlin (2020), Optimizing interim analysis timing for Bayesian adaptive commensurate designs, Statistics in Medicine, doi: 10.1002/sim.8414

Sibin Yang, Dao-Xin Yao and Anders W. Sandvik (2020), Deconfined quantum criticality in spin-1/2 chains with long-range interactions, arXiv: 2001.02821 [physics.comp-ph]

Muni Zhou, Nuno F. Loureiro, Dmitri A. Uzdensky (2020), Multi-scale dynamics of magnetic flux tubes and inverse magnetic energy transfer, arXiv: 2001.07291 [astro-ph.HE]

Catherine Zucker, Joshua S. Speagle, Edward F. Schlafly, Gregory M. Green, Douglas P. Finkbeiner, Alyssa Goodman, João Alves (2020), A compendium of distances to molecular clouds in the Star Formation Handbook, arXiv: 2001.00591 [astro-ph.GA]

Do you have news about research using computing resources at the MGHPCC? If you have an interesting project that you want to tell people about or a paper you would like listed, contact hlh@mit.edu

Links

MGHPCC Publications

Zeroing in on Decarbonization

Doctoral Candidate in MIT’s Department of Nuclear Science and Engineering (NSE) Nestor Sepulveda is using MGHPCC research computing resources to help chart a path towards decarbonization.

Read this story at MIT News

To avoid the most destructive consequences of climate change, the world’s electric energy systems must stop producing carbon by 2050. It seems like an overwhelming technological, political, and economic challenge — but not to Nestor Sepulveda.

“My work has shown me that we do have the means to tackle the problem, and we can start now,” he says. “I am optimistic.”

Sepulveda’s research, first as a master’s student and now as a doctoral candidate in the MIT Department of Nuclear Science and Engineering (NSE), involves complex simulations that describe potential pathways to decarbonization. In work published last year in the journal Joule, Sepulveda and his co-authors made a powerful case for using a mix of renewable and “firm” electricity sources, such as nuclear energy, as the least costly, and most likely, route to a low- or no-carbon grid.

These insights, which flow from a unique computational framework blending optimization and data science, operations research, and policy methodologies, have attracted interest from The New York Times and The Economist, as well as from such notable players in the energy arena as Bill Gates. For Sepulveda, the attention could not come at a more vital moment.

“Right now, people are at extremes: on the one hand worrying that steps to address climate change might weaken the economy, and on the other advocating a Green New Deal to transform the economy that depends solely on solar, wind, and battery storage,” he says. “I think my data-based work can help bridge the gap and enable people to find a middle point where they can have a conversation.”

An optimization tool

The computational model Sepulveda is developing to generate this data, the centerpiece of his dissertation research, was sparked by classroom experiences at the start of his NSE master’s degree.

“In courses like Nuclear Technology and Society [22.16], which covered the benefits and risks of nuclear energy, I saw that some people believed the solution for climate change was definitely nuclear, while others said it was wind or solar,” he says. “I began wondering how to determine the value of different technologies.”

Recognizing that “absolutes exist in people’s minds, but not in reality,” Sepulveda sought to develop a tool that might yield an optimal solution to the decarbonization question. His inaugural effort in modeling focused on weighing the advantages of utilizing advanced nuclear reactor designs against exclusive use of existing light-water reactor technology in the decarbonization effort.

“I showed that in spite of their increased costs, advanced reactors proved more valuable to achieving the low-carbon transition than conventional reactor technology alone,” he says. This research formed the basis of Sepulveda’s master’s thesis in 2016, for a degree spanning NSE and the Technology and Policy Program. It also informed the MIT Energy Initiative’s report, “The Future of Nuclear Energy in a Carbon-Constrained World.”

The right stuff

Sepulveda comes to the climate challenge armed with a lifelong commitment to service, an appetite for problem-solving, and grit. Born in Santiago, he enlisted in the Chilean navy, completing his high school and college education at the national naval academy.

“Chile has natural disasters every year, and the defense forces are the ones that jump in to help people, which I found really attractive,” he says. He opted for the most difficult academic specialty, electrical engineering, over combat and weaponry. Early in his career, the climate change issue struck him, he says, and for his senior project, he designed a ship powered by hydrogen fuel cells.

After he graduated, the Chilean navy rewarded his performance with major responsibilities in the fleet, including outfitting a $100 million amphibious ship intended for moving marines and for providing emergency relief services. But Sepulveda was anxious to focus fully on sustainable energy, and petitioned the navy to allow him to pursue a master’s at MIT in 2014.

It was while conducting research for this degree that Sepulveda confronted a life-altering health crisis: a heart defect that led to open-heart surgery. “People told me to take time off and wait another year to finish my degree,” he recalls. Instead, he decided to press on: “I was deep into ideas about decarbonization, which I found really fulfilling.”

After graduating in 2016, he returned to naval life in Chile, but “couldn’t stop thinking about the potential of informing energy policy around the world and making a long-lasting impact,” he says. “Every day, looking in the mirror, I saw the big scar on my chest that reminded me to do something bigger with my life, or at least try.”

Convinced that he could play a significant role in addressing the critical carbon problem if he continued his MIT education, Sepulveda successfully petitioned naval superiors to sanction his return to Cambridge, Massachusetts.

Simulating the energy transition

Since resuming studies here in 2018, Sepulveda has wasted little time. He is focused on refining his modeling tool to play out the potential impacts and costs of increasingly complex energy technology scenarios on achieving deep decarbonization. This has meant rapidly acquiring knowledge in fields such as economics, math, and law.

“The navy gave me discipline, and MIT gave me flexibility of mind — how to look at problems from different angles,” he says.

With mentors and collaborators such as Associate Provost and Japan Steel Industry Professor Richard Lester and MIT Sloan School of Management professors Juan Pablo Vielma and Christopher Knittel, Sepulveda has been tweaking his models. His simulations, which can involve more than 1,000 scenarios, factor in existing and emerging technologies, uncertainties such as the possible emergence of fusion energy, and different regional constraints, to identify optimal investment strategies for low-carbon systems and to determine what pathways generate the most cost-effective solutions.

“The idea isn’t to say we need this many solar farms or nuclear plants, but to look at the trends and value the future impact of technologies for climate change, so we can focus money on those with the highest impact, and generate policies that push harder on those,” he says.

Sepulveda hopes his models won’t just lead the way to decarbonization, but do so in a way that minimizes social costs. “I come from a developing nation, where there are other problems like health care and education, so my goal is to achieve a pathway that leaves resources to address these other issues.”

As he refines his computations with the help of MIT’s massive computing clusters, Sepulveda has been building a life in the United States. He has found a vibrant Chilean community at MIT and discovered local opportunities for venturing out on the water, such as summer sailing on the Charles.

After graduation, he plans to leverage his modeling tool for the public benefit, through direct interactions with policy makers (U.S. congressional staffers have already begun to reach out to him), and with businesses looking to bend their strategies toward a zero-carbon future.

It is a future that weighs even more heavily on him these days: Sepulveda is expecting his first child. “Right now, we’re buying stuff for the baby, but my mind keeps going into algorithmic mode,” he says. “I’m so immersed in decarbonization that I sometimes dream about it.”

Story image: Nestor Sepulveda (photo credit: Gretchen Ertl)

Related Publication

Nestor A. Sepulveda, Jesse D. Jenkins, Fernando J. de Sisternes, Richard K. Lester (2018), The Role of Firm Low-Carbon Electricity Resources in Deep Decarbonization of Power Generation, Joule, doi: 10.1016/j.joule.2018.08.006

Professor discovers way to differentiate individual black holes

Astrophysicists at UMass Dartmouth use computing resources at the MGHPCC to study black hole’s hair!
Read this story at UMass Dartmouth News

The black holes of Einstein’s theory of relativity can be described by three parameters: their mass, spin angular momentum, and electric charge. Since two extreme black holes that share these parameters cannot be distinguished, regardless of how they were made, black holes are said to “have no hair”: they have no additional attributes that can be used to tell them apart.

Physics Professor Gaurav Khanna and Professor Lior Burko of Georgia Gwinnett College recently published a paper in Physical Review Research that highlights how they are able to take measurements to discover a black hole’s hair.

Khanna and Burko used very intensive numerical simulations to generate their results. The simulations involved using dozens of the highest-end Nvidia graphics-processing-units (GPUs) with over 5,000 cores each, in parallel. “Each of these GPUs can perform as many as 7 trillion calculations per second; however, even with such computational capacity the simulations look many weeks to complete,” said Khanna.

The team showed that for extreme black holes, hair is a transient behavior. At first, they behave as extreme black holes would, but eventually they behave as regular, non-extreme black holes do. Burko summarized the result, saying, “Nearly extreme black holes that attempt to regrow hair will lose it and become bald again.” The team also discusses the observational feature with gravitational waves observatories such as LIGO/VIRGO or LISA which found the smoking-gun detection of nearly extreme black holes.

Story image: Artist rendition of the capture and tidal disruption of a star by a massive black hole via UMass Dartmouth News

Publication

Lior M. Burko, Gaurav Khanna, and Subir Sabharwal (2019), Transient scalar hair for nearly extreme black holes, Phys. Rev. Research, doi: 10.1103/PhysRevResearch.1.033106

Links

University of Massachusetts Research Computing

New model helps pave the way to bringing clean fusion energy down to Earth

Turbulence — the unruly swirling of fluid and air that mixes coffee and cream and can rattle airplanes in flight — causes heat loss that weakens efforts to reproduce on Earth the fusion that powers the sun and stars. Now scientists have modeled a key source of the turbulence found in a fusion experiment at the U.S. Department of Energy’s (DOE) Princeton Plasma Physics Laboratory (PPPL), paving the way for improving similar experiments to capture and control fusion energy. The research used computer resources at the MGHPCC.

Read this story at DOE Science News

State of the art simulations

The research, led by Juan Ruiz Ruiz while a graduate student at the Massachusetts Institute of Technology (MIT) and working with PPPL researchers Walter Guttenfelder and Yang Ren, used state-of-the-art simulations to zero-in on the source of the turbulence that produces heat loss. The findings predicted results consistent with experiments on the National Spherical Torus Experiment (NSTX) fusion device at PPPL, pinpointing the source as microscopic turbulent eddies. Driving these eddies is the gradient, or variation, in the electron temperature in the core of the plasma, the so-called electron temperature gradient (ETG).

Fusion combines light elements in the form of plasma — the hot, electrically charged state of matter composed of free electrons and atomic nuclei — that generates massive amounts of energy. Scientists around the world are seeking to replicate fusion on Earth for a virtually inexhaustible supply of power to generate electricity.

The recent findings confirmed theories of when ETG can be a main driver of electron turbulence, known as electron thermal transport, that whips up the heat loss in spherical tokamaks such as NSTX.  The consistency of the simulation with experimental data gives confidence “that the simulation contains the necessary physics to explain the loss of heat,” said Ruiz Ruiz, now a postdoctoral research assistant at the University of Oxford and first author of a paper reporting the results in Plasma Physics and Controlled Fusion.

The results apply to a type of H-mode, or high-confinement, experiment on the spherical NSTX, which is shaped more like a cored apple than the doughnut-like shape of more widely used conventional tokamaks. Understanding the source of electron thermal transport is a top priority for confining heat in future fusion facilities, and particularly in spherical tokamaks, which lose most of their heat through such transport in high-performance H-mode plasmas.

A little like radar

Ruiz Ruiz reached his conclusion by simulating a diagnostic called “high-k scattering” that NSTX researchers used to measure turbulence in the experiment. The technique scatters microwave radiation into the plasma, with the scattered radiation carrying information about the turbulence in the core. The process works a little like radar or sonar, Ruiz Ruiz says. The radiation bounces off objects — plasma eddies the size of electron orbits in this case — and reflects back their movement and positions.

Comparing the simulated and measured data called for painstakingly filtering and analyzing the vast output produced by the simulation code that Ruiz Ruiz used. “It takes a huge amount of effort to do an apples-to-apples comparison of the measured and simulated turbulence,” said Guttenfelder, who co-advised Ruiz Ruiz with an MIT professor. “Juan did just about the most thorough job you could do to show that the model is consistent with all the experimental data.”

Similar methods could be used to confirm the source of heat loss on the upgraded NSTX, called the NSTX-U, and the Mega Ampere Spherical Tokamak (MAST) in the United Kingdom, Ruiz Ruiz said. “That could demonstrate the ability of the simulations to accurately forecast the loss of heat — and therefore the performance — of spherical tokamaks,” he said.

Support for this work comes from the DOE Office of Science. Computer simulations were conducted at the National Energy Research Scientific Computing Center (NERSC), a DOE Office of Science user facility at the Lawrence Berkeley National Laboratory, and Massachusetts Green High Performance Computing Center, operated jointly by five Massachusetts research universities.

PPPL, on Princeton University’s Forrestal Campus in Plainsboro, N.J., is devoted to creating new knowledge about the physics of plasmas — ultra-hot, charged gases — and to developing practical solutions for the creation of fusion energy. The Laboratory is managed by the University for the U.S. Department of Energy’s Office of Science, which is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit energy.gov/science.

Publication:

J Ruiz Ruiz, W Guttenfelder, A E White, N T Howard, J Candy, Y Ren, D R Smith, N F Loureiro, C Holland and C W Domier (2019), Validation of gyrokinetic simulations of a National Spherical Torus eXperiment H-mode plasma and comparisons with a high-k scattering synthetic diagnostic, Plasma Physics and Controlled Fusion, doi: 10.1088/1361-6587/ab4742

Story Image: National Spherical Torus Experiment – Wikipedia

The Giant in our Stars

Harvard astronomers using computers housed at the MGHPCC discover largest known coherent gaseous structure in our galaxy.

Astronomers at Harvard University have discovered a monolithic, wave-shaped gaseous structure – the largest ever seen in our galaxy – made up of interconnected stellar nurseries. Dubbed the “Radcliffe Wave” in honor of the collaboration’s home base, the Radcliffe Institute for Advanced Study, the discovery transforms a 150-year-old vision of nearby stellar nurseries as an expanding ring into one featuring an undulating, star-forming filament that reaches trillions of miles above and below the galactic disk.

The work, published in Nature, was enabled by a new analysis of data from the European Space Agency’s Gaia spacecraft, launched in 2013 with the mission of precisely measuring the position, distance, and motion of the stars. The research team’s innovative approach combined the super-accurate data from Gaia with other measurements to construct a detailed, 3D map of interstellar matter in the Milky Way, and noticed an unexpected pattern in the spiral arm closest to Earth.

Computations were enabled using Harvard University’s Odyssey computing cluster housed at the MGHPCC.

Read more at the Harvard Gazette

Story image: In this illustration, the “Radcliffe Wave” data is overlaid on an image of the Milky Way galaxy. Image from the WorldWide Telescope, courtesy of Alyssa Goodman via The Harvard Gazette.

Related Publications:

João Alves, Catherine Zucker, Alyssa A. Goodman, Joshua S. Speagle, Stefan Meingast, Thomas Robitaille, Douglas P. Finkbeiner, Edward F. Schlafly and Gregory M. Green (2020), A Galactic-scale gas wave in the solar neighborhood, Nature, doi: 10.1038/s41586-019-1874-z

Catherine Zucker, Joshua S. Speagle, Edward F. Schlafly, Gregory M. Green, Douglas P. Finkbeiner, Alyssa Goodman, João Alves (2020), A compendium of distances to molecular clouds in the Star Formation Handbook, arXiv: 2001.00591 [astro-ph.GA]

Links

Harvard University Research Computing

December Publications

Below is a selection of papers that appeared in December 2019 reporting the results of research using the Massachusetts Green High Performance Computing Center (MGHPCC), or acknowledging the use of Harvard’s Odyssey Cluster, Northeastern’s Discovery Cluster, the Boston University Shared Computing Cluster and MIT’s Engaging Cluster all of which are housed at the MGHPCC.

Manuel A. Buen-Abad, Raymond T. Co, and Keisuke Harigaya (2019), Common Origin of Warm Dark Matter and Dark Radiation, arXiv:1911.13267 [astro-ph.CO]

Qian Di et al (2019), Assessing NO2 Concentration and Model Uncertainty with High Spatiotemporal Resolution across the Contiguous United States Using Ensemble Model Averaging, Environ. Sci. Technol., doi: 10.1021/acs.est.9b03358

Francesca Dominici, Joel Schwartz, Qian Di,Danielle Braun, Christine Choirat, and Antonella Zanobetti (2019), Assessing Adverse Health Effects of Long-Term Exposure to Low Levels of Ambient Air Pollution: Phase 1, Research Report 200. Boston, MA:Health Effects Institute .

Robyn S. Lee PhD, Jean-François Proulx MD, Fiona McIntosh BSc, Marcel A. Behr MD, William P. Hanage PhD (2019),  Investigating within-host diversity of Mycobacterium tuberculosis reveals novel super spreaders in the Canadian North, biorXiv, doi: 10.1101/801308

Tim Menke et al (2019), Automated discovery of superconducting circuits and its application to 4-local coupler design, arXiv: 1912.0332 [quant-ph]

Julian B. Munoz, Cora Dvorkin, Francis-Yan Cyr-Racine (2019), Probing the Small-Scale Matter Power Spectrum with Large-Scale 21-cm Data, arXiv: 1911.11144 [astro-ph.CO]

E. Pavadai, M.J. Rynkiewicz, A. Ghosh, W. Lehman (2019), Docking Troponin-T onto the Tropomyosin Overlapping Domain of Thin Filaments, Biophysical Journal, doi: 10.1016/j.bpj.2019.11.3393

Oliver H E Philcox, Daniel J Eisenstein (2019), Computing the Small-Scale Galaxy Power Spectrum and Bispectrum in Configuration-Space, Monthly Notices of the Royal Astronomical Society, doi: 10.1093/mnras/stz3335

R. Graham Reynolds et al (2019), Phylogeographic and phenotypic outcomes of brown anole colonization across the Caribbean provide insight into the beginning stages of an adaptive radiation, Journal of Evolutionary Biology, doi: 10.1111/jeb.13581

SaiLavanyaa Sundar, Michael J. Rynkiewicz, Anita Ghosh, William Lehman, and Jeffrey R. Moore (2019), Cardiomyopathy Mutation Alters End-to-end Junction of Tropomyosin and Reduces Calcium Sensitivity, Biophysical Journal, doi: 10.1016/j.bpj.2019.11.3396

Steven B. Torrisi, Arunima K. Singh, Joseph H. Montoya, Tathagata Biswas and Kristin A. Persson (2019), Two-Dimensional Forms of Robust CO2 Reduction Photocatalysts, arXiv: 1912.09545 [cond-mat.mtrl-sci]

Do you have news about research using computing resources at the MGHPCC? If you have an interesting project that you want to tell people about or a paper you would like listed, contact hlh@mit.edu

Links

MGHPCC Publications

Holyoke Coalition Initiates Clean Energy Transition Plan, aided by $400,000 in Grants

Holyoke, MA — A coalition comprised of the City of Holyoke, Neighbor to Neighbor Massachusetts and the Conservation Law Foundation are launching a planning initiative that aims to completely transition Holyoke’s buildings and energy grid away from fossil fuels.

Read this press release at the City of Holyoke website

The effort is made possible thanks to generous grant funding from the Barr Foundation, which earlier this month awarded the City of Holyoke with $275,000 and Neighbor to Neighbor with $125,000. The funding will cover costs related to project management, technical consulting, resident engagement and capacity building, and will provide additional technical support to the Holyoke Gas & Electric. Additionally, the Conservation Law Foundation will be contributing its time and resources to the coalition to enhance stakeholder engagement and help in the design of energy policies and approaches.

“On Earth Day this year, I spoke of the need for bold, collective action at the local level to confront our climate crisis and build a green economy. Over the past few months we’ve been building partnerships to launch an initiative that will allow Holyoke to lead the way in a world remade and powered by renewable energy, while serving as a model for cities around the country” said Holyoke Mayor Alex Morse. “I am grateful to the Barr Foundation for their funding support and thrilled to work with Neighbor to Neighbor and the Conservation Law Foundation as institutional partners in this endeavor.”

“We are grateful to the Barr Foundation for giving us the opportunity to make a grand shift in Holyoke’s energy future” added Elvis Méndez, Organizing Director for Neighbor to Neighbor. “Neighbor to Neighbor is ready and excited to work with the City of Holyoke to ensure that this transition is equitable and centers the community residents of Holyoke in the process.”

“Holyoke is such an exciting example of what’s possible when City leaders, community groups, and residents work together to seize the opportunity of clean energy,” said Mariella Puerto, Barr Foundation Co-Director of Climate. “The city now hosts one of the largest solar and energy storage facilities in the Commonwealth – where a coal-fired power plant once stood. The Barr Foundation is pleased to support this next phase in Holyoke’s efforts to achieve a prosperous and resilient community, free of fossil fuels.”

“With the climate crisis upon us, it’s imperative that we break our addiction to fossil fuels,” said CLF President Brad Campbell. “Holyoke’s bold action to ditch dirty fossil fuels and build the community’s future around clean energy sets the standard for cities and towns across New England to follow.”

In addition to the project partners, the Rocky Mountain Institute (RMI) has also pledged no-cost technical assistance to the Holyoke Gas & Electric to support initial analytical work by the municipal utility company. “RMI is excited to be working with the ambitious and talented team at HG&E,” said Bruce Nilles, Managing Director at RMI. “They are offering a bold and sustainable vision that can both help residents of Holyoke see a brighter and cleaner future, as well as inspire other cities to learn from their experiences.”

“Holyoke Gas & Electric is committed to providing the highest quality service and energy to our customers,” said Kate Sullivan Craven, HG&E’s Director of Marketing and Communications. “We are proud of our continuous record of innovation when it comes to clean energy and are looking forward to working with the Rocky Mountain Institute to further those initiatives.”

“Completing this work comes with great economic opportunity, whether it’s outfitting our building stock with modern heating systems, more energy-efficient envelopes or being at the cutting edge of clean energy technology,” said Marcos Marrero, the City’s Director of Planning & Economic Development. “We’re grateful to have such experienced partners to work with shoulder-to-shoulder in laying the path towards a clean energy future.”

The planning effort is set to begin early in 2020 by developing a project governance structure that includes a cross-section of community members and government officials, preparation for public engagement actions, and aggregating technical data.

The Barr Foundation’s mission is to invest in human, natural, and creative potential, serving as thoughtful stewards and catalysts. Based in Boston, Barr works to elevate the arts, advance solutions for climate change, and help all young people succeed in learning, work, and life. The Foundation’s Clean Energy efforts aim to accelerate a massive scale-up of renewables and energy efficiency across the Northeast. For more information, visit www.barrfoundation.org.

Contact: Stefan Lanfer, slanfer@barrfoundation.org Neighbor to Neighbor of Massachusetts (N2N) is a grassroots peoples organization made up of people of color, immigrants, women, and the working class, organizing to put people and the planet before profit. N2N seeks to build capacity and community power to transform the institutions that govern our lives. In an era of income inequality, environmental degradation, and racism, N2N chapters are building the power to confront this triple crisis in Massachusetts. N2N seeds new alternatives that put power and decision-making in the hands of those directly affected. For more information, visit https://www.n2nma.org/.

Contact: Elvis Méndez, Organizing Director, elvis@n2nma.org & Andrea Nyamekye, Campaign, and Policy Director andrea@n2nma.org

The Conservation Law Foundation (CLF) protects New England’s environment for the benefit of all people. We use the law, science, and the market to create solutions that preserve natural resources, build healthy communities, and sustain a vibrant regional economy. CLF’s approach to environmental advocacy is distinguished by our close involvement with local communities; our ability to design and implement effective strategies; and our capacity for developing innovative and economically sound solutions to our region’s most critical environmental challenges. For more information, visit https://www.clf.org/

Contact: Jake O’Neill, Press Secretary, joneill@clf.org

The Rocky Mountain Institute (RMI)’s mission is to transform global energy use to create a clean, prosperous, and secure low-carbon future. RMI engages businesses, communities, institutions, and entrepreneurs to accelerate the adoption of market-based solutions that cost-effectively shift from fossil fuels to efficiency and renewables. RMI is helping cities, communities, states, and regions meet their energy and climate goals, boost economic growth, and achieve the goals set out in the Paris Accord. For more information, visit https://rmi.org/

Contact: Bruce Niles, Managing Director, bnilles@rmi.org

The Holyoke Gas & Electric (HG&E) is a municipally-owned utility company which provides electric, natural gas and telecom solutions to residential and business customers in Holyoke and Southampton. The HG&E boasts an electric portfolio composed of over 50% renewable energy sources, and that is about 85% carbon-free. The utility continues to expand its renewable energy portfolio while maintaining some of the lowest electric rates in Massachusetts and New England. For more information, visit https://www.hged.com/

Contact: Kate Sullivan Craven, Director of Marketing and Communications, ksullivan@hged.com

MGHPCC @ Supercomputing19

This year, SC19, the International Conference for High-Performance Computing, Networking, Storage and Analysis,  was held at the Colorado Convention Center in Denver, November 17 – 22.  MGHPCC representation included the Boston University Research Computing booth which spotlighted several of the compute-intensive projects it has housed at the Center while Northeast Cyberteam project lead Julie Ma led a Birds of a Feather session providing an update on Ask.CI, the Q&A Platform for Research Computing.

Boston University’s booth at SC19  highlighting  the rich diversity of  projects supported by BU Research Computing Services and the multiple collaborative projects including the Northeast Storage Exchange (NESE), ATLAS NET2, Mass Open Cloud, the Northeast Cyber Team Program, Campus champions, CASC and CaRCC in which they are involved – image credit: Charles Jahnke

Launched in 2018, Ask.CI aggregates answers to a broad spectrum of questions that are commonly asked by the research computing community, creating a shared, archived, publicly-searchable knowledge base. “Establishing a Q&A site of this nature requires some tenacity,” says Ma. “While Ask.CI has gained traction in the year since its launch, attracting nearly 150,000 page views, hundreds of contributors, and worldwide participation, we are always seeking ways to grow our audience.”

With this in mind, Ask.CI recently introduced “locales“, institution-specific subcategories where institutions/communities of practice can post FAQs relevant to their constituents. The BoF session  provided representatives from Locales pilot participants Aaron Culich (University of California, Berkeley), Torey Battelle (Colorado School of Mines), John Goodhue (MGHPCC), Katia Oleinik and Jacob Pessin (Boston University), Vanessa Sochat (Stanford University), Dana Brunson (Internet2), Chris Hill (MIT), Thomas Cheatham III (University of Utah), and Zoe Braiterman (Open Web Applications Security Project, OWASP) an opportunity to share experiences and discuss future plans.

Story image courtesy J. Goodhue.

Related

Cluster Racing at SC18 MGHPCC News

“Ask” Q&A for Scientific Researchers to be Demonstrated at Prestigious SC18 Conference MGHPCC Press Release

The Fast and the Furious MGHPCC News

Holyoke Codes Maker Jam

Holyoke Codes presented Maker Jam at the MGHPCC on Saturday, November 23, a five-hour event celebrating technology, science, art, craft, creativity, and curiosity!

In association with Holyoke Media and The Print Shop, and with food generously sponsored by NERD (New England Regional Developers), the event included workshops, talks, and an interactive playground.

Among the workshops were:

Drones! Where visitors could try their hand flying a drone, learning how to control it with code, as well as program their own autonomous flights to master avoiding obstacles, and record photos and video.

Gliders: A CNC Machine for Aeronautics Education where representatives from AeroForm covered the development of their CNC (computer numerical control) machine, demonstrated its capabilities and provided a brief introduction to the science of flight with the construction of a free-flight glider.

The AeroForm is a tabletop CNC machine that cuts precision airfoils and other aircraft parts from solid polystyrene foam. Designed and manufactured in the Pioneer Valley, this unique machine brings research-level manufacturing capability to STEM classrooms, research labs, and home workshops.

 

Visitors get hands-on in the “Inventing with Micro:bit (and Hummingbird”) workshop.

Inventing with Micro-robotics Kits Micro:bit (and Hummingbird ) where attendees got a primer in making and programming with the block-based MakeCode as well as JavaScript and PythonMake, making things come alive with lights, sensors, and motors and a variety of programming languages

3D Model and Print Jewelry where visitors learned how to create 3D models with TinkerCAD an easy-to-use free tool for creating digital designs that are ready to be 3D printed into physical objects

The interactive playground included:

Holyoke Media, Holyoke’s public access media organization, offered visitors the chance to try GoPro cameras, record their own podcasts and experiment with green screen video.

Holyoke Media green screen.

 

The Print Shop, a local collaborative workspace that provides community access to commercial printing tools, production and classroom spaces, brought their 3-D sublimation printer for visitors to try out.

Soccer Robots invited visitors to try their hands at programing a Micro:bit robot soccer team controlled via radio using a computer vision system.

Exploring soccer robots.

 

Autonomous Race Cars demonstrated self-directing race cars powered by Raspberry Pi and machine learning with a Tensorflow 
neural network and a Deepracer with a reinforcement learning model.

FM Location Service allowed visitors to try and determine their geographic location by triangulating FM radio signals.

Mesh Networks and Sensors was an installation teaching how particle micro-controllers can be used to easily set up a mesh network for monitoring sensors in remote locations.

Internet-Enabled Gameboy demonstrated the fun to be had on a Gameboy with internet access.

Drive a Robot with Streaming Video illustrated how robots, controlled by an app built with a new App Inventor extension, enables people to build their own apps with streaming video.

Interactive Video App introduced a project with multitouch interactive video feedback, processing in real-time on webGL.

A visitor learns about the inputs to multitouch interactive video.

 

Bipedal Robots introduced an open-source, 3D printed, bipedal robot.

Visitors try our virtual reality.

Virtual Reality allowed visitors to experience and explore the world of VR.

Members of FLL and FTC robotics teams demonstrated the Tetrix and EV3 robots they have been creating for the FIRST robotics challenges. Related News

HolyokeCodes: Soccer Robots MGHPCC News

Teaching Computer Science Using Minecraft CSTA W Mass News

Holyoke Codes Cyber Security Workshop with Girls Inc. MGHPCC News

Links

HolyokeCodes

 

UMass-BU-Northeastern Team Receive NSF Grant to Develop New Cloud Computing Platforms

A new cloud computing testbed is coming to MGHPCC thanks to a $5M grant from NSF.

A team of researchers co-led by Mike Zink of UMass, Orran Krieger of BU, and Peter Desnoyers of NU have received an award from the National Science Foundation to construct and support a testbed for research and experimentation into new cloud platforms – the underlying software which provides cloud services to applications. Testbeds such as this are critical for enabling research into new cloud technologies. This is research that requires experiments which potentially can change the operation of the cloud itself. By providing capabilities that currently are only available to researchers within a few large commercial cloud providers, the new testbed will allow diverse communities to exploit these technologies, thus “democratizing” cloud computing research and allowing increased collaboration between the research and open source communities.

University Reporting

UMass Amherst Engineer Michael Zink Leads Research Team Developing New Cloud Computing Platforms UMass News

Collaboration Awarded an NSF Grant of $5M to Create New Cloud Computing Testbed BU News

The Future of Computing is a Better, Faster Cloud Northeastern News

Links

Mass Open Cloud

November Publications

Below is a selection of papers that appeared in October 2019 reporting the results of research using the Massachusetts Green High Performance Computing Center (MGHPCC), or acknowledging the use of Harvard’s Odyssey Cluster, Northeastern’s Discovery Cluster, the Boston University Shared Computing Cluster and MIT’s Engaging Cluster all of which are housed at the MGHPCC.

Mengxi Chen, Lin Hu,  Ashwin Ramasubramaniam, and  Dimitrios Maroudas (2019), Effects of pore morphology and pore edge termination on the mechanical behavior of graphene nanomeshes, Journal of Applied Physics, doi: 10.1063/1.5125107

Xueyan Feng, Christopher J. Burke, Mujin Zhuo, Hua Guo, Kaiqi Yang, Abhiram Reddy, Ishan Prasad, Rong-Ming Ho, Apostolos Avgeropoulos, Gregory M. Grason and Edwin L. Thomas (2019), Seeing mesoatomic distortions in soft-matter crystals of a double-gyroid block copolymer, Nature, doi: 10.1038/s41586-019-1706-1

Brandt Gaches (2019), The Impact of Protostellar Feedback on Astrochemistry, Doctoral Dissertation – University of Massachusetts Amherst, https://scholarworks.umass.edu/dissertations_2/1721

Aaron T. Lee, Stella S.R. Offner, Kaillin M. Kratter, Rachel A. Smullen, and Pak Shing Li (2019), The Formation and Evolution of Wide-orbit Stellar Multiples in Magnetized Clouds, arXiv:1911.07863 [astro-ph.GA]

D. Kirk Lewis and Sahar Sharifzadeh (2019), Defect-induced exciton localization in bulk gallium nitride from many-body perturbation theory, Physical Review Materials, doi: 10.1103/PhysRevMaterials.3.114601

Xiaorong Liu and Jianhan Chen (2020), Modulation of p53 Transactivation Domain Conformations by Ligand Binding and
Cancer-Associated Mutations, Conference Proc. Pacific Symposium on Biocomputing 25:195, https://psb.stanford.edu/psb-online/proceedings/psb20/Liu_X.pdf

Clara Maurel, Patrick Michel, J. Michael Owen, Richard P. Binzel, Megan Bruck-Syal, G. Libourel (2019), Simulations of high-velocity impacts on metal in preparation for the Psyche mission, Icarus, doi: 10.1016/j.icarus.2019.113505

Lindsay J. Underhill, Chad W. Milando, Jonathan I. Levy, W. Stuart Dols, Sharon K. Lee, M. Patricia Fabian (2019), Simulation of indoor and outdoor air quality and health impacts following installation of energy-efficient retrofits in a multifamily housing unit, Building and Environment, doi: 10.1016/j.buildenv.2019.106507

Aristoula Selevoua, George Papamokosa, Tolga Yildirimc, Hatice Duran, Martin Steinhart and George Floudas (2019), Eutectic liquid crystal mixture E7 in nanoporous alumina. Effects of confinement on the thermal and concentration fluctuations, RSC Adv., doi: 10.1039/C9RA08806G

Soplata, Austin Edward (2019), A Thalamocortical Theory of Propofol Phase-amplitude Coupling, Doctoral Dissertation – Boston University School of Medicine, https://search.proquest.com/openview/2e231ecca119628be0bdf376b72bc920/1

Do you have news about research using computing resources at the MGHPCC? If you have an interesting project that you want to tell people about or a paper you would like listed, contact hlh@mit.edu

Links

MGHPCC Publications

UMASS Researcher Receives NSF Grant for GPU-Enabled HPC Cluster at MGHPCC

GPU facilities will be made available to researchers through Internet2 links and regional computing partnerships at MGHPCC.Read this story at umass.edu

To support a broadly shared Graphic Processing Unit (GPU)-enabled high-performance computing cluster for the Institute for Applied Sciences (IALS), computational biophysicist Jianhan Chen, chemistry and biochemistry and molecular biology, with others, recently was awarded a two-year, $415,000 grant from the National Science Foundation (NSF) that will fill what Chen calls “a critical need” for enabling computation-intensive research activities on campus.

Although the UMass system has a traditional shared cluster housed at the Massachusetts Green High-performance Computing Center (MGHPCC) in Holyoke, Chen points out, the current cluster has “minimal GPU capacity” and the campus and IALS need dedicated GPU computing hardware to support their research communities. His co-principal investigators on the project are Erin Conlon, mathematics and statistics, Peng Bai, chemical engineering, Chungwen Liang, IALS director of computational modeling, and Matthew Moore, food science.

“When we put in the grant we solicited comments and surveyed the need from IALS and identified 30 labs that could use it,” Chen explains. “They testified to the need and committed to the cost-share with NSF, which will come from IALS, the College of Natural Sciences, College of Engineering, central IT and the Vice Chancellor for Research and Engagement. This is going to be a really unique entity on campus, and it will have a far-reaching impact,” he predicts. “It will be busy from the get-go.”

“I think NSF saw how much need and support we have. I want to particularly highlight the important contributions of Chris Misra and John Griffin of IT,” he adds. “They have taken the leadership in providing technical support that’s absolutely critical to me and other principal investigators on campus. Without them and their excellent help, this will not work, period.”

The new cluster, once carefully built up by Griffin, Chen and his co-investigators will be managed by the IALS Computational and Modeling Core to provide long-term stability for operation and management, serving 250 IALS-affiliated research labs across 27 departments and seven colleges. “The GPU facility offers high-speed single- and double-precision operations as well as extreme parallelism to enhance current activities that contribute to the open science movement,” project leaders state.

It will also contribute to efforts to integrate regional education, outreach, diversity, and economic activities, as the GPU facilities will be made available to researchers through Internet2 links and regional computing partnerships at MGHPCC. The researchers predict that the new cluster “will most likely lead to new developments and discoveries including novel GPU-enabled modeling and simulation technologies that may elucidate molecular mechanism of drug delivery, computational design catalysts for renewable energy and chemical synthesis, advanced computational analysis tools for next-generation informatics and big data, and improved understanding of risk and resistance to breast cancer.”

 

Story image: Helen Hill

Scaling HPC Education

With the explosion in artificial intelligence and machine learning, modeling, simulation, and data analytics, High Performance Computing (HPC) has grown to become an essential tool across academic disciplines. However, HPC expertise remains in short supply with a shortage of people who know how to make HPC systems work and how to use them. At September’s IEEE HPEC 2019 conference, a session chaired by Dr Julie Mullen (MIT LLSC) and Lauren Milechen (MIT EAPS) (who are involved with the MGHPCC hosted MIT SuperCloud System) provided a platform for members of local area research computing teams to share how they are scaling up HPC education in response.

In her presentation Julie Ma (Project Lead, Northeast Cyberteam Initiative, MGHPCC) presented “Northeast Cyberteam: A Workforce Development Strategy for Research Computing” describing activity within the Northeast Cyberteam Initiative, an NSF funded effort, now in its third and final year, to increase the effective use of cyberinfrastructure by researchers and educators at small and mid-sized institutions in Northern New England by making it easier to obtain support from expert Research Computing Facilitators outside of their immediate academic networks.

“Our Northeast Cyberteam Research Computing Facilitators combine technical knowledge and strong interpersonal skills with a service mindset and use their connections with cyberinfrastructure providers to ensure that researchers and educators have access to the best available resources,” Ma explains. “It is widely recognized that such facilitators are critical to successful utilization of cyberinfrastructure, but in very short supply. The Northeast Cyberteam aims to build a pool of Research Computing Facilitators in the region and a process to share them across institutional boundaries. At the same time, we are providing experiential learning opportunities for students interested in becoming Research Computing Facilitators, as well as developing a self-service learning toolkit to provide timely access to information when it is needed.”

Mullen and Milechen, who are both intimately involved with the day-to-day running of the MIT SuperCloud System, used their presentation to describe the development and ongoing progress of a MOOC for teaching how to write scalable code through the use of standard workflows and a SPOC (Small, Private, Online Course) for training on the specifics of using and running on the MIT SuperCloud System.

“Most HPC centers recognize the need to provide their users with HPC training however, the limited time and resources available make this training and education difficult to scale to a growing and broadening audience. MOOCs (Massine Open Online Courses) can provide more accessible and scalable learning paths toward HPC expertise. In our talk, we presented MOOCs and their related technologies and teaching approaches, outlining how MOOC courses differ from face-to-face training, video-capturing of live events, webinars, and other established teaching methods with respect to pedagogical design, development issues, and deployment concerns,” says Milechen.

Robert Freeman directs Research Technology Operations at Harvard Business School (HBS). His talk “Humans in Scaling Research Computing Facilitation and Education” again focused on the challenge of growing the specialized workforce needed to respond to accelerating growth in campus research computing.

“Scaling people-efforts in HPC facilitation and education is an important problem as science and research programs are no longer isolated, work-in-silos efforts; and increasing complexity on all fronts drives an increased need for a better-trained workforce for both research and support staff,” says Freeman. “A number of communities, both local and national, are working on these efforts using multiple approaches. In my talk I discussed specific themes, highlighting the institutions and organizations (both historical and ongoing) that play a part, that have met success and encourage participation, and all of which are growing opportunities to democratize and evangelize these ever-changing advanced cyberinfrastructure resources: creating communities in education, bringing HPC/HTC (high throughput computing) to all disciplines, bringing facilitation approaches to everyone, and building communities for enabling research.”

In particular, Freeman drew attention to the Campus Research Computing Consortium (CARCC) an organization seeking to develop, advocate for, and advance campus research computing and data and associated professions in response to the accelerating rate of change in the area encouraging his audience to perhaps contribute to CARCC efforts by themselves helping enrich the consortium.

Brian Gregor is a member of the Research Computing Staff (RCS) at Boston University. He used his talk “Developing HPC Skills Across the University Community” to share the experience of the BU RCS’s eight-member Applications Support team who work with the > 2,000 researchers across the university using BU’s Shared Computing Cluster.

“We teach tutorials at the start of each semester on a variety of programming topics from introductions to Linux and cluster programming to advanced programming in R and Python,” said Gregor. “In 2018 our tutorials had approximately 1200 attendees. Tutorial attendance continues to grow in 2019 with demand especially high for our set of Python tutorials.”

“Over the past four years,” he continued, “the team has become increasingly involved in teaching specialized topics for academic classes including cluster usage, HPC software, big data tools such as Spark, and other programming languages. In the academic arena, we assist in areas that include deep learning, computational biomedicine, and biostatistics, as well as a graduate data science program in the department of mathematics. As with our tutorials, the interest in our teaching at the academic level only continues to grow with each passing semester.”

To accommodate the increased demand for HPC skills education the BU team plans rollout of OnDemand for easier access to the cluster for academic classes and the research community.

In response to the increasing demand for HPC skills education and training, he said his team was developing video versions of their introductory tutorials to help meet the increasing demand and to free up time to introduce tutorials on more advanced topics. He also said that his team was starting an internship program for graduate students interested in improving their HPC skills and in learning about research facilitation.

Related

HPEC ’19 MGHPCC News

Lincoln Laboratory’s new artificial intelligence supercomputer is the most powerful at a university

TX-GAIA is tailor-made for crunching through deep neural network operations.

Read this story at MIT News

The new TX-GAIA (Green AI Accelerator) computing system at the Lincoln Laboratory Supercomputing Center (LLSC) has been ranked as the most powerful artificial intelligence supercomputer at any university in the world. The ranking comes from TOP500, which publishes a list of the top supercomputers in various categories biannually. The system, which was built by Hewlett Packard Enterprise, combines traditional high-performance computing hardware — nearly 900 Intel processors — with hardware optimized for AI applications — 900 Nvidia graphics processing unit (GPU) accelerators.

“We are thrilled by the opportunity to enable researchers across Lincoln and MIT to achieve incredible scientific and engineering breakthroughs,” says Jeremy Kepner, a Lincoln Laboratory fellow who heads the LLSC. “TX-GAIA will play a large role in supporting AI, physical simulation, and data analysis across all laboratory missions.”

TOP500 rankings are based on a LINPACK Benchmark, which is a measure of a system’s floating-point computing power, or how fast a computer solves a dense system of linear equations. TX-GAIA’s TOP500 benchmark performance is 3.9 quadrillion floating-point operations per second, or petaflops (though since the ranking was announced in June 2019, Hewlett Packard Enterprise has updated the system’s benchmark to 4.725 petaflops). The June TOP500 benchmark performance places the system No. 1 in the Northeast, No. 20 in the United States, and No. 51 in the world for supercomputing power. The system’s peak performance is more than 6 petaflops.

But more notably, TX-GAIA has a peak performance of 100 AI petaflops, which makes it No. 1 for AI flops at any university in the world. An AI flop is a measure of how fast a computer can perform deep neural network (DNN) operations. DNNs are a class of AI algorithms that learn to recognize patterns in huge amounts of data. This ability has given rise to “AI miracles,” as Kepner puts it, in speech recognition and computer vision; the technology is what allows Amazon’s Alexa to understand questions and self-driving cars to recognize objects in their surroundings. The more complex these DNNs grow, the longer it takes for them to process the massive datasets they learn from. TX-GAIA’s Nvidia GPU accelerators are specially designed for performing these DNN operations quickly.

TX-GAIA is housed in a new modular data center, called an EcoPOD, at the LLSC’s green, hydroelectrically powered site in Holyoke, Massachusetts. It joins the ranks of other powerful systems at the LLSC, such as the TX-E1, which supports collaborations with the MIT campus and other institutions, and TX-Green, which is currently ranked 490th on the TOP500 list.

Kepner says that the system’s integration into the LLSC will be completely transparent to users when it comes online this fall. “The only thing users should see is that many of their computations will be dramatically faster,” he says.

Among its AI applications, TX-GAIA will be tapped for training machine learning algorithms, including those that use DNNs. It will more quickly crunch through terabytes of data — for example, hundreds of thousands of images or years’ worth of speech samples — to teach these algorithms to figure out solutions on their own. The system’s compute power will also expedite simulations and data analysis. These capabilities will support projects across the laboratory’s R&D areas, such as improving weather forecasting, accelerating medical data analysis, building autonomous systems, designing synthetic DNA, and developing new materials and devices.

TX-GAIA, which is also ranked the No. 1 system in the U.S. Department of Defense, will also support the recently announced MIT-Air Force AI Accelerator. The partnership will combine the expertise and resources of MIT, including those at the LLSC, and the U.S. Air Force to conduct fundamental research directed at enabling rapid prototyping, scaling, and application of AI algorithms and systems.

Story image: TX-GAIA is housed inside of a new EcoPOD, manufactured by Hewlett Packard Enterprise, at the site of the Lincoln Laboratory Supercomputing Center in Holyoke, Massachusetts. Photo: Glen Cooper

 

 

Supercomputer analyzes web traffic across entire internet

Researchers at the Lincoln Laboratory Supercomputing Center use the MIT SuperCloud to model web traffic potentially aiding cybersecurity, computing infrastructure design, Internet policy, and more.

Read this story at MIT News

Using a supercomputing system, MIT researchers have developed a model that captures what web traffic looks like around the world on a given day, which can be used as a measurement tool for internet research and many other applications.

Understanding web traffic patterns at such a large scale, the researchers say, is useful for informing internet policy, identifying and preventing outages, defending against cyberattacks, and designing more efficient computing infrastructure. A paper describing the approach was presented at the recent IEEE High Performance Extreme Computing Conference (HPEC 2019).

For their work, the researchers gathered the largest publicly available internet traffic dataset, comprising 50 billion data packets exchanged in different locations across the globe over a period of several years.

They ran the data through a novel “neural network” pipeline operating across 10,000 processors of the MIT SuperCloud, a system that combines computing resources from the MIT Lincoln Laboratory and across the Institute. That pipeline automatically trained a model that captures the relationship for all links in the dataset — from common pings to giants like Google and Facebook, to rare links that only briefly connect yet seem to have some impact on web traffic.

The model can take any massive network dataset and generate some statistical measurements about how all connections in the network affect each other. That can be used to reveal insights about peer-to-peer filesharing, nefarious IP addresses and spamming behavior, the distribution of attacks in critical sectors, and traffic bottlenecks to better allocate computing resources and keep data flowing.

In concept, the work is similar to measuring the cosmic microwave background of space, the near-uniform radio waves traveling around our universe that have been an important source of information to study phenomena in outer space. “We built an accurate model for measuring the background of the virtual universe of the Internet,” says Jeremy Kepner, a researcher at the MIT Lincoln Laboratory Supercomputing Center and an astronomer by training. “If you want to detect any variance or anomalies, you have to have a good model of the background.”

Joining Kepner on the paper are: Kenjiro Cho of the Internet Initiative Japan; KC Claffy of the Center for Applied Internet Data Analysis at the University of California at San Diego; Vijay Gadepally and Peter Michaleas of Lincoln Laboratory’s Supercomputing Center; and Lauren Milechin, a researcher in MIT’s Department of Earth, Atmospheric and Planetary Sciences.

Breaking up data

In internet research, experts study anomalies in web traffic that may indicate, for instance, cyber threats. To do so, it helps to first understand what normal traffic looks like. But capturing that has remained challenging. Traditional “traffic-analysis” models can only analyze small samples of data packets exchanged between sources and destinations limited by location. That reduces the model’s accuracy.

The researchers weren’t specifically looking to tackle this traffic-analysis issue. But they had been developing new techniques that could be used on the MIT SuperCloud to process massive network matrices. Internet traffic was the perfect test case.

Networks are usually studied in the form of graphs, with actors represented by nodes, and links representing connections between the nodes. With internet traffic, the nodes vary in sizes and location. Large supernodes are popular hubs, such as Google or Facebook. Leaf nodes spread out from that supernode and have multiple connections to each other and the supernode. Located outside that “core” of supernodes and leaf nodes are isolated nodes and links, which connect to each other only rarely.

Capturing the full extent of those graphs is infeasible for traditional models. “You can’t touch that data without access to a supercomputer,” Kepner says.

In partnership with the Widely Integrated Distributed Environment (WIDE) project, founded by several Japanese universities, and the Center for Applied Internet Data Analysis (CAIDA), in California, the MIT researchers captured the world’s largest packet-capture dataset for internet traffic. The anonymized dataset contains nearly 50 billion unique source and destination data points between consumers and various apps and services during random days across various locations over Japan and the U.S., dating back to 2015.

Before they could train any model on that data, they needed to do some extensive preprocessing. To do so, they utilized software they created previously, called Dynamic Distributed Dimensional Data Mode (D4M), which uses some averaging techniques to efficiently compute and sort “hypersparse data” that contains far more empty space than data points. The researchers broke the data into units of about 100,000 packets across 10,000 MIT SuperCloud processors. This generated more compact matrices of billions of rows and columns of interactions between sources and destinations.

Capturing outliers

But the vast majority of cells in this hypersparse dataset were still empty. To process the matrices, the team ran a neural network on the same 10,000 cores. Behind the scenes, a trial-and-error technique started fitting models to the entirety of the data, creating a probability distribution of potentially accurate models.

Then, it used a modified error-correction technique to further refine the parameters of each model to capture as much data as possible. Traditionally, error-correcting techniques in machine learning will try to reduce the significance of any outlying data in order to make the model fit a normal probability distribution, which makes it more accurate overall. But the researchers used some math tricks to ensure the model still saw all outlying data — such as isolated links — as significant to the overall measurements.

In the end, the neural network essentially generates a simple model, with only two parameters, that describes the internet traffic dataset, “from really popular nodes to isolated nodes, and the complete spectrum of everything in between,” Kepner says.

Using supercomputing resources to efficiently process a “firehose stream of traffic” to identify meaningful patterns and web activity is “groundbreaking” work, says David Bader, a distinguished professor of computer science and director of the Institute for Data Science at the New Jersey Institute of Technology. “A grand challenge in cybersecurity is to understand the global-scale trends in Internet traffic for purposes, such as detecting nefarious sources, identifying significant flow aggregation, and vaccinating against computer viruses. [This research group has] successfully tackled this problem and presented deep analysis of global network traffic,” he says.

The researchers are now reaching out to the scientific community to find their next application for the model. Experts, for instance, could examine the significance of the isolated links the researchers found in their experiments that are rare but seem to impact web traffic in the core nodes.

Beyond the internet, the neural network pipeline can be used to analyze any hypersparse network, such as biological and social networks. “We’ve now given the scientific community a fantastic tool for people who want to build more robust networks or detect anomalies of networks,” Kepner says. “Those anomalies can be just normal behaviors of what users do, or it could be people doing things you don’t want.”

 

Story image:

Using a supercomputing system, MIT researchers developed a model that captures what global web traffic could look like on a given day, including previously unseen isolated links (left) that rarely connect but seem to impact core web traffic (right).

Image courtesy of the researchers, edited by MIT News

Related

HPEC’19 MGHPCC News

The MGHPCC Supercloud

October Publications

Below is a selection of papers that appeared in October 2019 reporting the results of research using the Massachusetts Green High Performance Computing Center (MGHPCC), or acknowledging the use of Harvard’s Odyssey Cluster, Northeastern’s Discovery Cluster, the Boston University Shared Computing Cluster and MIT’s Engaging Cluster all of which are housed at the MGHPCC.

Connor Bottrell, Maan H. Hani, Hossen Teimoorinia, Sara L. Ellison, Jorge Moreno, Paul Torrey, Christopher C. Hayward, Mallory Thorp, Luc Simard and Lars Hernquist (2019), Deep learning predictions of galaxy merger stage and the importance of observational realism, arXiv: 1910.07031 [astro-ph.GA]

Xiang Chen, Juan Ruiz Ruiz, Nathan Howard, Walter Guttenfelder, Jeff Candy, Jerry Hughes, Robert Granetz, Anne White
(2019),  Prediction of high-k electron temperature fluctuation in an NSTX H-mode plasma, abstract for 61st Annual Meeting of the APS Division of Plasma Physics,
http://meetings.aps.org/Meeting/DPP19/Session/GP10.158

Cedric Flamant, Grigory Kolesov, Efstratios Manousakis, Efthimios Kaxiras (2019), Imaginary-Time Time-Dependent Density Functional Theory and Its Application for Robust Convergence of Electronic States, J. Chem. Theory Comput., doi: 10.1021/acs.jctc.9b00617

Peng Liang and Juan Pablo Trelles (2019), 3D numerical investigation of a free-burning argon arc with metal electrodes using a novel sheath coupling procedure, Plasma Sources Science and Technology, doi: 10.1088/1361-6595/ab4bb6

Peilong Li, Chen Xu, Hao Jin, Chunyang Hu, Yan Luo, Yu Cao, Jomol Mathew, Yunsheng Ma (2019), ChainSDI: A Software-Defined Infrastructure for Regulation-Compliant Home-Based Healthcare Services Secured by Blockchains, IEEE Systems Journal, doi: 10.1109/JSYST.2019.2937930

Philip L Pagano, Qi Guo, Chethya Ranasinghe, Evan Schroeder, Kevin Robben, Florian Häse, Hepeng Ye, Kyle Wickersham, Alán Aspuru-Guzik, Dan T. Major, Lokesh Gakhar, Amnon Kohen, Christopher M. Cheatum (2019), Oscillatory Active-site Motions Correlate with Kinetic Isotope Effects in Formate Dehydrogenase, ACS Catal., doi: 10.1021/acscatal.9b03345

Pranay Patil and Anders W. Sandvik (2019), Hilbert Space Fragmentation and Ashkin-Teller Criticalityin Fluctuation Coupled Ising Models, arXiv: 1910.03714 [cond-mat.str-el]

Juan Ruiz Ruiz (2019), Validation of gyrokinetic simulations in NSTX including comparisons with a synthetic diagnostic for high-k scattering, abstract for 61st Annual Meeting of the APS Division of Plasma Physics, http://meetings.aps.org/Meeting/DPP19/Session/TI2.1

Debjani Sihi, Eric A. Davidson, Kathleen E. Savage, Dong Liang (2019), Simultaneous numerical representation of soil microsite production and consumption of carbon dioxide, methane, and nitrous oxide using probability distribution functions, Global Change Biology, doi: 10.1111/gcb.14855

Nia S. Walker, Rosa Fernández, Jennifer M. Sneed, Valerie J. Paul, Gonzalo Giribet, David Combosch (2019), Differential Gene Expression during Substrate Probing in Larvae of the Caribbean Coral, Molecular Ecology, doi: 10.1111/mec.15265

Cheng-Chiang Wu, Fay-Wei Li, Elena M. Kramer (2019), Large-scale phylogenomic analysis suggests three ancient superclades of the WUSCHEL-RELATED HOMEOBOX transcription factor family in plants, PLoS ONE, doi: 10.1371/journal.pone.0223521

Do you have news about research using computing resources at the MGHPCC? If you have an interesting project that you want to tell people about or a paper you would like listed, contact hlh@mit.edu

Links

MGHPCC Publications

The Computer Will See You Now

Vijaya B. Kolachalama, Ph.D., is an Assistant Professor at the Boston University School of Medicine. His area of expertise is in computational biomedicine and in particular machine learning and computer vision.

Activity in the Kolachalama Lab falls into two broad categories: machine learning and computer vision for precision medicine, and research into device-artery interactions, interfacial mechanics and drug delivery. In his machine learning work, Kolachalama makes extensive use of BU’s Shared Computing Cluster housed at the MGHPCC.

“Artificial intelligence is poised to help deliver precision medicine, yet achieving this goal is nontrivial,” says Kolachalama. “Machine learning and image processing techniques along with developments in software and hardware technologies allow us to consider questions across a range of scales,” he continues. “In the radiology and digital pathology work in my lab, we leverage these tools for pattern recognition and understanding pathophysiological mechanisms, paving the way for the development of new and, we hope, more effective and accessible, diagnostic and prognostic biomedical technologies geared to a range of diseases.”

Vijaya Kolachalama is an Assistant Professor of Medicine at Boston University’s Medical School. In his work, he uses BU’s Shared Computing Cluster,  housed at the MGHPCC, in research seeking to apply machine learning and computer vision for precision medicine.

A 2018 paper about his use of deep neural networks to help in the assessment of chronic kidney disease (Kolachalama, 2018) exemplifies his lab’s approach and methodologies, applying advanced machine learning techniques to systematize digital pathology.

“Chronic kidney damage is routinely assessed semiquantitatively by scoring the amount of disease seen in a renal biopsy sample,” explains Kolachalama. “Although image digitization and morphometric techniques have made quantifying the extent of damage easier, the advanced machine learning tools we are developing provide a more systematic way to stratify kidney disease severity.”

Speaking to BU News at the time, Kolachalama said, that “While the trained eyes of expert pathologists are able to gauge the severity of disease and detect nuances of kidney damage with remarkable accuracy, such expertise is not available in all locations, especially at a global level”. Recognizing the potential of his team’s model to act as a surrogate nephropathologist, especially in resource-limited settings, Kolachalama noted that “If healthcare providers around the world can have the ability to classify kidney biopsy images with the accuracy of a nephropathologist right at the point-of-care, then this can significantly impact practice.”

More recently, Kolachalama has applied his machine learning techniques similarly in other areas including Alzheimer’s disease, and osteoarthritis.

“It remains difficult to characterize the source of pain in knee joints either using radiographs or magnetic resonance imaging (MRI),” he explains. “In work with Gary Chang (Chang et al, 2019), a Postdoctoral Associate in my lab, we were interested to see if using deep neural networks could distinguish knees with pain from those without it as well as to perhaps identify the structural features that are associated with knee pain.”

In that study, the team constructed a convolutional Siamese network to associate MRI scans obtained on subjects from the NIH’s Osteoarthritis Initiative with frequent unilateral knee pain, comparing the knee with frequent pain to the contralateral knee without pain in order to map model-predicted regions of high pain association. An expert radiologist then compared the MRI scans with the derived maps to identify the presence of abnormalities.

The radiologist’s review revealed that about 86% of the cases that were predicted correctly had effusion-synovitis within the regions that were most associated with pain, suggesting deep learning can be applied to assess knee pain from MRI scans.

In the context of Alzheimer’s disease, in a study, this time working with Shangran Qiu, a graduate student in the Physics Department at BU, Kolachalama and co-authors applied their machine learning tools to explore whether by combining MRI data with results from the Mini–Mental State Examination (MMSE) and logical memory tests the accuracy of diagnosing mild cognitive impairment could be enhanced (Qui et al, 2019.)

“We combined deep learning models trained on MRI slices to generate a fused MRI model using different voting techniques to predict normal cognition versus mild impairment. We then combined the fused MRI model with a second class of deep learning models trained on data obtained from NIH’s National Alzheimer Coordinating Center database containing individuals with normal cognition and mild cognitive impairment,” Kolachalama explains. “Our fused model did better than the individual models alone with an overall accuracy of over 90%”

Finally, in a collaboration between researchers in the Kolachalama Lab with researchers at Visterra Inc, a clinical-stage biotechnology company committed to developing innovative antibody-based therapies for the treatment of patients with kidney diseases and other hard-to-treat diseases, Kolachalama was recently involved in a study published in the journal Protein Engineering, Design & Selection (Wollacott et al, 2019), applying his machine-learning tools to quantify the “nativeness” of antibody sequences.

“Antibodies can be useful in treating, in particular, cancer and autoimmune diseases, and it has been shown that synthetic antibodies that more closely resemble their natural counterparts demonstrate improved rates of expression and stability,” explains Kolachalama. “Antibodies often undergo substantial engineering en route to the generation of a therapeutic candidate with good developability properties. Characterization of antibody libraries has shown that retaining native-like sequence improves the overall quality of the library. Using a bi-directional long short-term memory (LSTM) network model to score sequences for their similarity to naturally occurring antibodies, we were able to demonstrate our model was able to outperform other approaches at distinguishing human antibodies from those of other species.”

“None of this work would be possible without access to BU’s Shared Computing Cluster and by extension the Massachusetts Green High Performance Computing Center in Holyoke where it is housed,” says Kolachalama. “Our access to them is indispensable in advancing our work towards developing clinically useful digital pathology tools.”

Story image: Trichrome-stained images from renal biopsy samples at different magnifications – image courtesy V. Kolachalama

Related Publications:

Kolachalama V.B., Singh P., Lin C.Q., Mun D., Belghasem M.E., Henderson J.M., Francis J.M., Salant D.J., Chitalia V.C.(2018), Association of Pathological Fibrosis with Renal Survival Using Deep Neural Networks, Kidney Int. Rep., doi: 10.1016/j.ekir.2017.11.002

Chang G.H., Felson D.T., Qiu S., Capellini T.D., Kolachalama V.B. (2019), Assessment of bilateral knee pain from MR imaging using deep neural networks, bioRxiv, doi: 10.1101/463497

Qiu, S., Chang G.H., Panagia M., Gopal D.M., Au R., Kolachalama V.B. (2018), Fusion of deep learning models of MRI scans, Mini–Mental State Examination, and logical memory test enhances diagnosis of mild cognitive impairment, Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, doi: 10.1016/j.dadm.2018.08.013

Andrew M Wollacott, Chonghua Xue, Qiuyuan Qin, June Hua, Tanggis Bohnuud, Karthik Viswanathan, Vijaya B Kolachalama (2019), Quantifying the nativeness of antibody sequences using long short-term memory networks, Protein Engineering, Design and Selection, doi: 10.1093/protein/gzz031

Links

Kolachalama Lab

Boston University Shared Computing Cluster

New AI Technology Significantly Improves Human Kidney Analysis BU News

Pages