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.
Exscalate4CoV consortium hits the ground running
Leveraging a new €3 million grant from the European Commission, a group of 18 institutions and research centers called Exscalate4CoV is attacking the coronavirus on a number of fronts. Exscalate4CoV is aiming to identify drugs and strategies for effect coronavirus treatment and containment, among other objectives. To do that, the program is leveraging the power of several supercomputing centers, including Italy’s CINECA, Germany’s Jülich and Spain’s Barcelona Supercomputing Center.
To read more, visit the press release here. You can also listen to This Week in HPC’s exclusive interview with CINECA’s Carlo Cavazzoni (embedded below) or read the transcript on HPCwire here.
Folding@home folds the coronavirus
The massive, crowdsourced computing platform Folding@home, which focuses on protein analysis and molecular dynamics simulations, is focusing its 100,000+ constituent volunteer machines (estimated at nearly 100 petaflops) on analysis of the viral proteins of the coronavirus. The consortium is aiming to score a repeat of its prior success with identifying drug targets for a protein in the ebola virus that was previously considered “undruggable.” To learn more about Folding@home’s work to fight the coronavirus – including how you can sign up to donate idle time on your own computer or gaming console – visit the HPCwire article here.
Graph databases help track the spread of COVID-19
A group of Chinese researchers leveraged Neo4j, an open-source graph database, to build an application that allows Chinese citizens to perform a self-assessment by entering places, flights, trains or license plates and returning all known cases associated with those entries, along with the “edges” or travel routes that connect them. Another graph database, which used Nebula Graph, zeroed in on examining how five people were infected with the coronavirus in Tianjin. To read more about both of these projects, visit the Datanami article here.
Insilico Medicine applies AI for coronavirus drug development
A Maryland-based company that uses AI for drug discovery, Insilico Medicine identified thousands of molecules that may be candidates for COVID-19 drugs after just a few days of applying 28 different machine learning models. Furthermore, Insilico has promised to “synthesize and test up to 100 molecules using its own resources and the resources generously offered by its closest partners,” with six completed at the time of their statement. To read more about Insilico’s work, visit the EnterpriseAI story here.
Summit crunches the “spike” protein
Oak Ridge National Laboratory (ORNL) has dedicated Summit, the world’s fastest publicly ranked supercomputer, to coronavirus research. Summit’s 148 Linpack petaflops are simulating the coronavirus’ “spike” protein, which researchers believe is key to its ability to infect. By testing how various compounds interact with that protein, the researchers hope they can find a way to disable the spike protein. Already, they have identified 77 candidate compounds that warrant further testing by medical researchers. To read more about the coronavirus research being conducted on Summit, visit the HPCwire article here.
AI- and supercomputer-powered CT scans aid with diagnosis
In China, COVID-19 has been primarily detected through its particular expression of pneumonia on CT scans. One Beijing startup, Infervision, deployed an AI-powered COVID-19 pneumonia detection tool at dozens of hospitals, helping to process tens of thousands of patients. For more on Infervision’s work, click visit the EnterpriseAI story here.
Similarly, Chinese researchers trained an AI model on China’s first petascale supercomputer, Tianhe-1, that helped to distinguish between COVID-19 and non-COVID-19 pneumonia on CT scans. The researchers reported that the trained model outperformed both test kits and human radiologists. To read more, visit the HPCwire story here.
The NSF and the European Commission open their doors – and wallets – for more computing research on the coronavirus
Major scientific funding organizations are issuing loud calls for new computing research on the coronavirus. The NSF issued two Dear Colleague Letters calling for “non-medical, non-clinical-care research” proposals for COVID-19, specifically highlighting computing activities through its Office of Advanced Cyberinfrastructure and offering supplemental funding.
The European Commission has announced an aggregate €47.5 in funding for COVID-19 research, including computing activities. €3 million of this has already gone to the supercomputing-powered Exscalate4CoV consortium, and the EU is highlighting prioritized and fast-tracked access for COVID-19 research at other European computing facilities.
To read more about these calls for research, visit the HPCwire story here.
Big data companies offer free access for COVID-19 researchers
Cloud file storage company Qumulo has announced that its software is available for free, effective immediately, to public and private research organizations that are working to fight the coronavirus. “Using Qumulo’s software to manage and understand petabytes of data real-time, medical and research organizations around the world can work together and leverage the power of the cloud and hybrid environments to fight COVID-19,” said Bill Richter, president and CEO of Qumulo. Similarly, analytics company Starburst is offering its enterprise Presto software for free and TigerGraph is allowing free use of its graph database technology to support COVID-19 research. To read more, visit the article on Datanami 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.