by Kevin Schofield
This week's "Weekend Reads" include a big, important breakthrough on protein folding, a look at how Washington nonprofits are responding to the COVID-19 crisis, and King County's analysis of where people are being exposed to COVID-19.
Protein Folding Almost Solved!
This week a group of researchers at Google announced that their "Deep Mind" computer system had made a huge leap forward in its ability to predict "protein folding," how protein sequences curl up onto themselves to form specific shapes.
Proteins are the stuff that life is created from: long sequences of amino acids that together make up cells and most of the machinery inside of them. Our DNA — which itself is composed of proteins — is basically a cookbook with detailed instructions for how to construct all the different pieces that make up a human body, mostly from proteins. A protein can have chemical properties that allow it to interact with other molecules — and other proteins — to achieve some function. But proteins also have structural properties: their shapes matter a great deal, as they determine what parts of the protein are exposed and what kind of physical interactions they can have. Essentially as soon as a protein is manufactured inside a cell it instantly curls up into a specific shape that is determined by its sequence of amino acids — and every duplicate of that protein folds into the same shape. But until now, scientists have been unable to predict that shape by reading the amino acids. For a typical protein, there might be millions of potential shapes that it might take based upon the combinations of chemical bonds between different parts as they come near each other.
The great paradox of molecular biology has been that despite the fact that a protein instantly folds up in a particular way, and always folds up the same way — it clearly knows what it's doing — it has been an insanely difficult problem to try to predict it. But doing so would be an enormous boost to medical research, allowing for new kinds of "designer drugs" and cancer medicines tailored to specific biological functions or carcinomas. There are about 180 million known protein sequences from the natural world, but only about 170,000 of them have known structures.
The Google team announced this week that its new system can predict protein folding within an atom-length for two thirds of the proteins it was tested with, and it was highly accurate for the other third. The system was trained on the 170,000 known protein structures, and as experimental results add more data to the database on specific proteins' structures, retraining the system will likely make it even more accurate.
If the system's results hold up broadly, this could represent a huge breakthrough moment for medical research.
How Nonprofits Are Responding to the COVID-19 Pandemic
It's been a tough year for nonprofits;across the board donations are down, and arts and cultural organizations have seen their other revenue sources dry up as shutdowns force them to keep their doors closed. A team of researchers at UW's Evans School of Public Policy just published a study on how Washington nonprofits have responded to the crisis.
It cites two big challenges for them this year. The first is funding, which has decreased substantially this year with the exception of government grant and contract funding; in fact, they report that program service funding is down almost 48%. The second big challenge is a drop in volunteers: the pandemic has made many volunteers skittish about coming in and performing work around others. And yet, demand for services is up this year for health and human services related nonprofits (it's down for other types of nonprofits).
The research team found that five common responses among nonprofits:
-Changing how programs are provided;
-Pausing some programs;
-Altering short-term goals;
-Changing priorities;
-Adding new programs to address emerging needs.
Organizations seemed to avoid ending programs, altering long-term organizational goals, changing the organizational mission, or ceasing operations entirely.
How and Where Are People in King County Catching COVID?
Last week the King County public health department released a study of how and where individuals in King County likely contracted COVID-19, based upon interviews that health department investigators did to track down outbreaks. The results are very interesting.
It's important to recognize that the county-wide response to the virus has evolved over time: recommendations changed and restrictions were imposed and eased in stages. The report looks at aggregate statistics since March, but also zooms in on the past sixty days to give a better idea of where the most common sources of new infections are today.
Over the past two months, the most common exposure settings were:
-Within a household (33.7% of new infections);
-Non-healthcare workplaces (20.7%);
-Community and social gatherings (17.5%);
-A reported close contact (13.2%).
Healthcare settings, travel, homeless shelters, and other congregate settings were remarkably low.
It's interesting to look at the trends over time too: in March, healthcare settings accounted for half of all infections, but that figure has dropped steadily over the months. Whereas household infections have grown steadily as a share of the total infections, with a big jump in November.
The report also breaks out the infection statistics by race/ethnicity, age, and area of the county.
At the moment, the moral of the story is that we need to pay more attention to hygiene in our own homes, and probably get more serious about protecting our "pods." Some are using this data, and similar reports in other jurisdictions, to argue for reducing restrictions on businesses. That's not a strong argument, however, as "community/social gatherings," which includes restaurants, trended much higher when restrictions were looser. And at the moment, none of the settings are the source for the majority of new cases — though the trend for households is worrisome.
Featured image: MedGlobal.Org under a Creative Commons 2.0 license.
Kevin Schofield is a freelance writer and the founder of Seattle City Council Insight, a website providing independent news and analysis of the Seattle City Council and City Hall. He also co-hosts the "Seattle News, Views and Brews" podcast with Brian Callanan, and appears from time to time on Converge Media and KUOW's Week in Review.
Before getting into journalism Kevin worked at Microsoft for 26 years, including 17 in the company's research division. He has twin daughters, loves to cook, and is trying hard to learn Spanish and the guitar.
Before you move on to the next story . . . please consider that the article you just read was made possible by the generous financial support of donors and sponsors. The Emerald is a BIPOC-led nonprofit news outlet with the mission of offering a wider lens of our region's most diverse, least affluent, and woefully under-reported communities. Please consider making a one-time gift or, better yet, joining our Rainmaker Family by becoming a monthly donor. Your support will help provide fair pay for our journalists and enable them to continue writing the important stories that offer relevant news, information, and analysis. Support the Emerald!
Before you move on to the next story …
The South Seattle Emerald™ is brought to you by Rainmakers. Rainmakers give recurring gifts at any amount. With around 1,000 Rainmakers, the Emerald™ is truly community-driven local media. Help us keep BIPOC-led media free and accessible.
If just half of our readers signed up to give $6 a month, we wouldn’t have to fundraise for the rest of the year. Small amounts make a difference.
We cannot do this work without you. Become a Rainmaker today!