For the second (and, hopefully, final) year in a row, SC21 included a second major research award alongside the ACM 2021 Gordon Bell Prize: the Gordon Bell Special Prize for High Performance Computing-Based COVID-19 Research. Last year, the first iteration of this award went to simulations of the SARS-CoV-2 spike protein; this year, the prize went to researchers at Japan’s RIKEN who simulated the dynamics of aerosolized COVID droplets, shaping behavior around the world at the height of the pandemic.
RIKEN’s Fugaku, which has topped the Top500 for four consecutive cycles – though perhaps with an asterisk this time – launched wildly ahead of schedule in 2020 in order to provide researchers with unprecedented resources as they began to run intensive COVID-19 research. Over the coming months, Fugaku and the researchers it hosted made specific names for themselves among the innumerable supercomputing organizations tackling COVID: they consistently focused on the propagation of COVID droplets, whether through masks, around face shields, across barriers or in the cold. To accomplish this, the RIKEN researchers built on top of industrial particle simulation software, optimizing and adding to the software to allow it to run quickly at scale on the massive Fugaku system.
That work shaped behavior both in Japan and around the world as our civilization struggled to navigate the uncertainties and anxieties of the pandemic. “What particularly impressed the committee was that this work changed public behavior in Japan and internationally in the early stages of the pandemic,” said Mark Parsons, chair of the prize committee.
The specific research awarded with this Gordon Bell Special Prize was titled Digital Transformation of Droplet/Aerosol Infection Risk Assessment Realized on Fugaku for the Fight against COVID-19 and was authored by six RIKEN researchers: Kazuto Ando, Rahul Bale, ChungGang Li, Satoshi Matsuoka, Keiji Onishi and Makoto Tsubokura.
The award – one of the few accepted in person at SC21’s awards ceremony – comes with a $10,000 prize courtesy of computing pioneer Gordon Bell.
To learn more about Fugaku’s COVID droplet simulations, read any of the below articles:
It’s Fugaku vs. COVID-19: How the World’s Top Supercomputer Is Shaping Our New Normal
RIKEN’s Ongoing COVID Research Includes New Vaccines, New Tests & More
At ISC, the Fight Against COVID-19 Took the Stage – and Yes, Fugaku Was There
ACM Gordon Bell Special Prize nominees
Of course, RIKEN’s droplet research is far from the only groundbreaking COVID research run on supercomputers – as detailed in HPCwire’s timeline of supercomputing and COVID, supercomputers around the world have devoted countless processing hours to hunting for therapeutics, studying epidemiology to stymie the spread of COVID and much, much more. RIKEN and Fugaku may have won the prize, but the other five nominees similarly exhibit some of the most impressive work done in humanity’s fight against the pandemic.
Language Models for the Prediction of SARS-CoV-2 Inhibitors
A team from Oak Ridge National Laboratory, for instance, trained the BERT deep learning language model on around 9.6 billion molecules, teaching it to generate and score COVID drug candidates based on protein binding affinity predictions. This work, run on Oak Ridge’s Summit supercomputer – still the most powerful publicly ranked system in the U.S. – reduced pre-training time from days to hours while greatly increasing the training dataset size.
Data-Driven Scalable Pipeline Using National Agent-Based Models for Real-Time Pandemic Response and Decision Support
At the University of Virginia, six researchers developed an “integrated, data-driven operational pipeline based on national agent-based models to support federal- and state-level pandemic planning and response.” The pipeline coordinates jobs across two HPC systems and collects, integrates and organizes COVID data across a number of governmental levels in order to power a digital twin of social interactions among 288 million individuals and billions of interactions across the United States.
#COVIDisAirborne: AI-Enabled Multiscale Computational Microscopy of Delta SARS-CoV-2 in a Respiratory Aerosol
This team of researchers hails from UC San Diego, the University of Illinois, the University of Pittsburgh, the University of Bristol, Argonne National Laboratory and Entos, Inc. – and it includes Rommie Amaro, lead of the research that won the Gordon Bell Special Prize in 2020. This year, their nominated research sought to “completely revise current models of airborne transmission of respiratory viruses by providing never-before-seen atomic-level views of the SARS-CoV-2 virus within a respiratory aerosol.”
FEP-Based Large-Scale Virtual Screening for Effective Drug Discovery against COVID-19
Another drug discovery screening project – this time led by a team from Sun Yat-sen University, the State Key Laboratory of HPC, Tianjin’s National Supercomputing Center, the Ocean University of China and the University of Kentucky – used a “rigorous and accurate method” for drug discovery called FEP-ABFE. They ran the screening across around 12,000 protein-ligand binding systems, powering the analysis with a new Tianhe supercomputer and identifying 50 compounds with “significant inhibitory activity[.]”
Intelligent Resolution: Integrating Cryo-EM with AI-Driven Multi-Resolution Simulations to Observe the SARS-CoV-2 Replication-Transcription Machinery in Action
The final nominated team, hailing from the University of Illinois, the California Institute of Technology, the University of Chicago, Argonne National Laboratory and the University of Leeds, shone a light on the the process by which SARS-CoV-2 replicates and transcribes the viral mRNA in human cells. To do this, they leveraged molecular dynamics simulations, AI and a workflow manager to handle the diverse workloads across HPC centers.
Congratulations to all the winners and nominees!