Mystery of Why Some People Don't Catch COVID

Would seem such a question could be sorted out with a machine learning approach. 

The mystery of why some people don’t catch COVID  in ArsTechnica

Scientists think they might hold the key to helping protect us all.

GRACE BROWNE, WIRED.COM

We all know a “COVID virgin,” or “Novid,” someone who has defied all logic in dodging the coronavirus. But beyond judicious caution, sheer luck, or a lack of friends, could the secret to these people’s immunity be found nestled in their genes? And could it hold the key to fighting the virus?

In the early days of the pandemic, a small, tight-knit community of scientists from around the world set up an international consortium, called the COVID Human Genetic Effort, whose goal was to search for a genetic explanation as to why some people were becoming severely sick with COVID while others got off with a mild case of the sniffles.

After a while, the group noticed that some people weren’t getting infected at all—despite repeated and intense exposures. The most intriguing cases were the partners of people who became really ill and ended up in intensive care. “We learned about a few spouses of those people that—despite taking care of their husband or wife, without having access to face masks—apparently did not contract infection,” says András Spaan, a clinical microbiologist at Rockefeller University in New York.

Spaan was tasked with setting up an arm of the project to investigate these seemingly immune individuals. But they had to find a good number of them first. So the team put out a paper in Nature Immunology in which they outlined their endeavor, with a discreet final line mentioning that “subjects from all over the world are welcome.”

The response, Spaan says, was overwhelming. “We literally received thousands of emails,” he says. The sheer volume rushing to sign up forced them to set up a multilingual online screening survey. So far, they’ve had about 15,000 applications from all over the world.

The theory that these people might have preexisting immunity is supported by historical examples. There are genetic mutations that confer natural immunity to HIV, norovirus, and a parasite that causes recurring malaria. Why would COVID be any different, the team rationalized? Yet, in the long history of immunology, the concept of inborn resistance against infection is a fairly new and esoteric one. Only a few scientists even take an interest. “It’s such a niche field, that even within the medical and research fields, it’s a bit pooh-poohed on,” says Donald Vinh, an associate professor in the Department of Medicine at McGill University in Canada. Geneticists don’t recognize it as proper genetics, nor immunologists as proper immunology, he says. This is despite there being a clear therapeutic goal. “If you can figure out why somebody cannot get infected, well, then you can figure out how to prevent people from getting infected,” says Vinh.  ... ' 

Making Decision Makers Use and Understand the Value of Models

Many times had to consider how to get key decision makers to use the results of analytical models.  This article touches on that in some ways. Like to consider further how this could be done consistently. 

Making machine learning more useful to high-stakes decision makers

A visual analytics tool helps child welfare specialists understand machine learning predictions that can assist them in screening cases.

Adam Zewe | MIT News Office

The U.S. Centers for Disease Control and Prevention estimates that one in seven children in the United States experienced abuse or neglect in the past year. Child protective services agencies around the nation receive a high number of reports each year (about 4.4 million in 2019) of alleged neglect or abuse. With so many cases, some agencies are implementing machine learning models to help child welfare specialists screen cases and determine which to recommend for further investigation.

But these models don’t do any good if the humans they are intended to help don’t understand or trust their outputs.

Researchers at MIT and elsewhere launched a research project to identify and tackle machine learning usability challenges in child welfare screening. In collaboration with a child welfare department in Colorado, the researchers studied how call screeners assess cases, with and without the help of machine learning predictions. Based on feedback from the call screeners, they designed a visual analytics tool that uses bar graphs to show how specific factors of a case contribute to the predicted risk that a child will be removed from their home within two years.... '