AI Talk: COVID-19 tracker, shots and tech titans tango

April 17th, 2020 / By V. “Juggy” Jagannathan, PhD

This week’s AI Talk…

Tracking the path of COVID-19

A fascinating new approach to tracking the COVID-19 virus migration during this pandemic comes from a software called Nextstrain. This tool confirmed that the virus took the same path Columbus took five centuries ago—from Europe to the Americas This is one of the first examples of a tool that enables what has been dubbed, genomic epidemiology. How does the tool work? As the name suggests, it uses gene sequencing. Essentially, as the virus transmits and spreads, it also mutates. The mutation of the genetic code is very slight—perhaps just one letter in its long code, but that is sufficient to paint a picture of where the genetic code came from. It is like a family tree where each code variant inherits from a parent node. If you couple that with the location information of where the sample genetic code was taken, you can paint a visual picture of the spread of the virus. You can look at the visualization created by Nextstrain here. Which features fascinating evolving pictures which are being continually updated. They can’t predict where the virus will jump to next, but it is instructive to see where the virus came from.

Vaccine shots based on genetic blueprint?

Curiously, I came across another gene-related story this week. The basic idea discussed in the article is how to use the genomic sequence of the COVID-19 virus to design a vaccine. These new genes are injected as vaccine and they trick the body into making a mutated and non-lethal form of the virus which triggers the immune response and antibody creation. The Coalition for Epidemic Preparedness Innovations (CEPI) approached a company working in this space, Inovio, in January of this year. The company downloaded the virus genetic sequences and managed to design a vaccine in three hours. It took longer to manufacture and start a clinical trial which happened last week. And, it turns out they are not alone. Apparently, this technology has been brewing for a few decades, but it looks like it might be ready for primetime now. It remains to be seen what will actually work. However, Bill Gates believes we should not wait to figure out which one of these different vaccines is going to work. He is helping fund seven new factories to produce potential vaccines at scale, knowing full well we are likely to waste billions of dollars in the process. Of course, speed to market at scale is of the essence and has the potential to save millions of lives and reboot the economy.

Tech titans tango

In order to return to normalcy following the pandemic, two factors have to be addressed. One is ability to test, the other is ability to do contact tracing. Let’s focus on the second part here. Basically, contact tracing is the process of finding each sick person and figuring out who they might have inadvertently compromised by being physically close to them. To address the contact tracing problem, Apple and Google announced this week that they are working together to address it. The tech titans plan to release interoperable APIs in early May that would allow public health agencies to develop contact tracing in a manner that is transparent and protects privacy. Detailed technical specs related to the underlying API framework, Bluetooth usage and security protocols have been released. How does this work? This article in The Verge does a pretty decent job of explaining the approach used. Briefly, everyone (assuming you want this protection) downloads an app to their phone. The phone periodically broadcasts an encoded identifier using Bluetooth, literally saying “I am here” and any other phone that is in the vicinity simply logs these IDs coming from nearby phones. When someone tests positive for the virus, they declare the diagnosis to a central database. All phones connect to the central database and periodically check whether any IDs in their local logs came from an infected person. If yes, they get a warning. That’s the whole scheme in a nutshell. Of course, there are layers of security and protection against hacking, compromising identities, etc. And, there is the matter of figuring out how to mimic the physical separation of six feet using Bluetooth. But, all in all, a pretty clever solution to the contact tracing problem. Let’s hope they role this out soon and we have an effective means of achieving this functionality.

Acknowledgement

My friend and former boss, Juergen Fritsch, sent me the info on the gene tracking effort.

I am always looking for feedback and if you would like me to cover a story, please let me know. “See something, say something”! Leave me a comment below or ask a question on my blogger profile page.

V. “Juggy” Jagannathan, PhD, is Director of Research for 3M M*Modal and is an AI Evangelist with four decades of experience in AI and Computer Science research.


During a pandemic, healthcare information is gathered, studied, and published rapidly by scientists, epidemiologists and public health experts without the usual processes of review. Our understanding is rapidly evolving and what we understand today will change over time. Definitive studies will be published long after the fact. 3M Inside Angle bloggers share our thoughts and expertise based on currently available information.