How AI Is Helping Humanity Tackle the Coronavirus Crisis

Like any tool, technology can be used for both good and bad. And sometimes, that bad is inadvertent; tech in the form of airplanes helped expedite the spread of the coronavirus around the world. But fortunately, technology will also aid in stopping this pandemic crisis.

A few weeks ago, we wrote about how the San Francisco-based company BlueDot utilized artificial intelligence (AI) to warn the general public about the dangers of COVID-19 well ahead of health officials. In case you missed it, you can read it here.

Examples like BlueDot give us hope that emerging tools like AI can help humanity tackle problems like the coronavirus in unprecedented ways. Critics of the technology often say it’s only tested in “sandbox scenarios” that offer limited significance for the real world. But that’s clearly not the case.

In this post, we’ll take a look at how two major AI organizations, DeepMind and IBM, are leveraging the smart technology to fight against the coronavirus.

IBM Strikes Back at the Coronavirus With Summit

The speed at which the novel coronavirus spreads presents an unprecedented challenge for the world. Consequently, researchers must accelerate their work in order to stymy this pandemic. Fortunately, Summit, the world’s fastest supercomputer, is here to help.

Solving the World’s Biggest Problems Is Summit’s Purpose

In 2014, New York-based developer IBM was commissioned by the US Department of Energy to build a supercomputer for a unique reason — to solve the most impactful problems facing the world. With the power of 200 petaflops and the capability to perform 200 quadrillion calculations per second, Summit is 1 million times more powerful than the fastest laptop in existence today.

At its home in Tennessee’s Oak Ridge National Laboratory, Summit has already helped tackle various monumental issues. The supercomputer has identified common cellular system patterns that precede Alzheimer’s disease, predicted extreme weather events via climate simulations, and analyzed contributory genes for traits like opioid addiction.

Identifying Crucial Chemical Compounds

Equipped with cutting-edge developments in artificial intelligence (AI), Summit was able to run thousands of simulations to elucidate which drug compounds have a good chance of stopping the coronavirus from infecting host cells.

To infect and compromise host cells, viruses inject them with a “spike” of genetic material. Finding chemicals that bind to this spike could potentially help stop the spread of the coronavirus. After creating a model of the coronavirus spike based on research published in January, Oak Ridge Lab research Micholas Smith then simulated how its particles would react to a myriad of compounds.

After over 8,000 simulations, Summit identified 77 promising compounds and even ranked them on likeliness to bind to the virus’ spike. Since more information on the coronavirus is coming out every day, the team will run simulations with Summit again to improve accuracy.

Still, the work that Oak Ridge Lab and Summit have accomplished so far is a great start. The next step would be to conduct experimental studies to validate the performance of these chemicals. This work could eventually lead to a greater understanding of how to improve our treatment of COVID-19.

How DeepMind Is Striving to Slow the Spread of Coronavirus

Perhaps known best for beating human players at a variety of games like Go and DOTA, London-based developer DeepMind is also working on solutions in the field of drug discovery. The Alphabet-owned company has been working on AlphaFold, a deep learning system that attempts to accurately predict protein structures — even when no similar ones exist.

Visualizing Solutions in 3D

The 3D structure of proteins is essential for crafting new medications, especially for new viruses. As we’ve alluded to, COVID-19 has spikey proteins jutting out from its surface; these are the spikes that compromise host cells. Lung cells contain an abundance of factors that allow these spikes to attach to them, which is why the coronavirus causes respiratory problems.

DeepMind’s AlphaFold is on a mission to figure out which proteins can thwart COVID-19’s attachment to host cells. As Summit did, AlphaFold is tapping into all of the newly available information about the coronavirus to do this. But rather than simulating thousands of different chemical compounds, AlphaFold is applying machine learning to focus on a few understudied proteins that carry immense potential for becoming drug or vaccine targets.

A Fundamental Problem In Drug Discovery

Protein folding is an enigmatic, decades-old issue in the fields of biochemistry and drug discovery. Nearly all our existing drug solutions latch onto certain proteins to function; identifying new protein structures for our fight against the coronavirus is much more complex.

But why is understanding how proteins fold so complex? Well, genetic code doesn’t seamlessly translate to how proteins look. And when it comes to new viruses, we’re basically fighting them blindfolded without predicting protein structures.

AlphaFold combines decades of deep learning advancements with pertinent information available from public domain protein structure databases. By using genome sequences, it can predict the properties of particular proteins. But it’s not like it can concoct specific sequences with special powers; it’s more like AlphaFold can offer us a quick police sketch of the “perpetrator.”

DeepMind emphasizes that “these structure predictions have not been experimentally verified.” It’s difficult to evaluate how and what AlphaFold will be able to contribute to helping humanity during this current pandemic. But it’s certainly en route towards aiding us in tackling future health crises.

AI’s Still No Substitute for Social Distancing

The coronavirus is bringing countries and economies to a standstill. So any help we can get is welcomed. It’s unlikely that AI will be the panacea we need to address this pandemic. Similar to the 2003 SARS crisis, our best defense against illnesses like these is one that we can each do without any newfangled technology: social distancing.

As with many other aspects of our society, COVID-19 is putting AI to the test. But with some hope, empathy, and teamwork, humanity will overcome this pandemic. For now, be safe, stay informed, and remember to wash your hands often.