Doctors and medical researchers have struggled to pinpoint let alone explain the deluge of symptoms induced by COVID-19 infections in patients, and what was once seen as a purely respiratory virus is beginning to be seen as a far more wide-ranging ailment. Now, new research from a team at Oak Ridge National Laboratory (ORNL) is using supercomputing power to illuminate how SARS-CoV-2 instigates various symptoms in hitherto unexplained ways.

The team, led by Dan Jacobson (chief scientist for computational systems biology at ORNL), compared the genes of cells from the lung fluids of nine COVID-19-infected patients to similar cells from 40 control patients. They were hunting for co-expression that is, correlation between certain genes being active or inactive. To comb through the genetic data, they turned to two of ORNLs HPC systems: Summit and Rhea.

Summits 4,608 nodes, each powered by two IBM Power9 CPUs and six Nvidia Volta GPUs, deliver 148 Linpack petaflops, placing it second on the most recent Top500 list of the worlds most powerful supercomputers. Rhea, meanwhile, is a 521-node cluster equipped with Intel Xeon E5 CPUs and Nvidia K80 GPUs that is well-suited for post-processing of data from more powerful systems.

Using this computing firepower, the researchers completed 2.5 billion correlation calculations over the course of a week. Then, they had what Jacobson describes as a eureka moment. In COVID-19-infected cells, the researchers found increased expression of enzymes that produce bradykinin (a compound that makes blood vessels dilated and permeable), decreased expression of enzymes that break down bradykinin and decreased expression of an enzyme that helps stall a catastrophic cascade: a bradykinin storm.

Based on their results, the team believes that a bradykinin storm may explain COVID-19s wide range of symptoms such as muscle pain, fatigue, headaches and brain fog better than the feared cytokine storm, a similar (and much better-known) effect where the body releases too many of the cytokine proteins that help regulate the human immune system.

We believe that when you take the inhibition at the top of this pathway off, you end up with an out-of-control cascade that leads to an opening up of the blood vessels, causing them to leak, Jacobson said in an interview with ORNLs Rachel Harken. If that happens in the lung, thats not good. Immune cells that are normally contained in the blood vessels flood into the surrounding infected tissue, causing inflammation.

The team is hopeful that if their results are validated by experimental analysis, at least ten different known drugs may prove promising to assist patients suffering from a bradykinin storm. If we can block this pathogenesis in severe patients, we can keep the human response from going overboard and give their immune system time to fight off the virus so they can recover, Jacobson said.

The researchers also found that the lung fluids of COVID-19 patients had higher expression of genes that increase the production and decrease the breakdown of hyaluronic acid a substance that can make patients feel like theyre trying to breathe through Jell-O. The team is hopeful that drug compounds known to treat this acid buildup will now also be explored.

To read ORNLs Rachel Harkens reporting on this research, click here.

See the original post:

Supercomputer-Powered Research Uncovers Signs of 'Bradykinin Storm' That May Explain COVID-19 Symptoms - HPCwire