Supercomputing, big data and artificial intelligence are crucial tools in the fight against the coronavirus pandemic. Around the world, researchers, corporations and governments are urgently devoting their computing resources to this global crisis. This column collects the biggest news about how advanced technologies are helping us fight back against COVID-19.
The King Abdullah University of Science and Technology (KAUST) Supercomputing Core Laboratory (KSL) has issued a call for COVID-19-related research proposals for its Shaheen II supercomputer (as well as other cluster computing resources). Shaheen II, which delivers 5.5 Linpack petaflops, placed 38th on the most recent Top500 list of the world’s most powerful publicly ranked supercomputers. To read more, click here.
Atos is highlighting its numerous initiatives to spur COVID-19 research and action, including a wide range of work by its supercomputers, a workplace services offering and participation in the “COVID-19 Dataset Challenge” competition that tasks researchers with applying machine learning to coronavirus questions. To read more, click here.
A team from Lawrence Berkeley National Laboratory has developed a tool called COVIDScholar that uses natural language processing and supercomputing to scan academic papers on COVID-19 and make the results easily searchable. Within a month, the team collected over 60,000 research papers, using automated scripts to pull the papers, standardize them and index them for searching. To read more, click here.
A duo of researchers from the University of Delaware have received an NSF grant to construct simulations of COVID-19 using supercomputers. The researchers will conduct molecular simulations at the atomic level using tools previously applied for successful analysis of AIDS and hepatitis B. To learn more, click here.
Earlier this year, two researchers from MIT combined Julia’s neural networks with differential equations to create “universal differential equations.” Now, those researchers have published a paper showing the use of universal differential equations for COVID-19 epidemiology, using to tool to illustrate how different approaches to quarantine and isolation measures affected the virus’ reproduction in Wuhan, Italy, South Korea and the U.S.
A team from the University of Alabama in Huntsville is using HPE’s Sentinel supercomputer, via Microsoft Azure, to engage in drug discovery for COVID-19. Sentinel, which is a Cray XC50 Intel-based system in Azure’s Texas datacenter, was used to investigate 20,000 molecules for their ability to disable proteins on COVID-19. To read more, click here.
Researchers at Penn State University are applying quantum machine learning, a cross between quantum information processing and machine learning, to assist with COVID-19 drug development. The researchers previously applied the same technique to solve similar combinatorial optimization problems. To read more, click here.
UK Research and Innovation (UKRI) has published a list of supercomputing resources that are available for researchers who want to conduct research on COVID-19. The list includes a wide range of resources across over a dozen facilities and institutions, all accessible via a single application. To read more, click here.
Researchers from the Texas Advanced Computing Center (TACC) combed through tens of millions of COVID-19-related tweets to provide researchers with a massive social media dataset, as well as some supercomputer-enabled insights. For now, those insights include trending words and phrases over time, but they will soon expand to include analyses of related terms, public figures and organization and a searchable database. To read more, click here.
Researchers at the University of Utah have used resources on TACC’s Frontera and Longhorn systems to rapidly construct molecular models of COVID-19-relevant compounds. The team generated more than 2,000 molecular models with the hopes of finding a new peptide inhibitor that can be experimentally verified in the coming weeks. To read more, click here.
Do you know about COVID-19 research that should be featured on this list? If so, send us an email at [email protected]. We look forward to hearing from you.