Supercomputers mimicking brain activity and looking for COVID treatments

October 15, 2021 – Machine learning has come a long way in the quarter of a century since a computer nicknamed Deep Blue shocked the world by defeating chess master Garry Kasparov. Today, with our smartphones having far more processing power than Deep Blue, scientists are targeting even bigger adversaries, including potentially deadly diseases like cancer, heart disease and COVID-19.

When supercomputers search for new drug cocktails to treat these diseases, scientists can feed the machines with mountains of data from decades of studies to aid analysis. But the coronavirus is still too new and mutating too quickly for scientists to turn to these common strategies.

Massachusetts Institute of Technology researchers have come up with a new way to address the lack of data on the new virus. They train computers to run algorithms that are modeled on the signal networks in the human brain. Like the brain, these neural networks can “learn” and adapt to rapidly changing information to create new connections in the blink of an eye.

In order to identify drug combinations that could work against COVID-19, the investigators ask their neural computer network to evaluate two things at the same time.

One of these is to look for drug pairs that, when taken together, are more powerful antivirals than either drug alone. This concept that two drugs are more effective together is known as “drug synergy”.

The computer also looks for parts of a disease that drugs are targeting, such as proteins or genetic mutations associated with a disease. The idea behind these two approaches is that the machines can “learn” which drug cocktails may have the greatest antiviral effect.

In their study,published in the Proceedings of the National Academy of Sciences, MIT scientists reveal two potential drug cocktails that they have found using this approach. One combines remdesivir, which has already been approved by the FDA for the treatment of COVID-19, and reserpine, a drug used for high blood pressure. The other combination is remdesivir and an experimental drug called IQ-1S, one of a family of drugs used to treat autoimmune diseases such as rheumatoid arthritis.

These drug cocktails have not yet been shown to be effective against COVID-19 in human studies. But the study results can help drug developers figure out which combinations make the most sense to test when looking for new treatments.