Sunday, November 15, 2020

THE QUOTE RACK

 Lasting T-cell immunity to Covid?

People who have recovered from Covid-19 may retain cellular immunity for at least six months after infection—a major boost for hopes of long-term protection against the virus. Recent research has found that the body’s initial antibody response to the disease diminishes quickly, suggesting antibodies won’t provide lasting immunity.

But so-called T-cells—a part of the immune system that attacks infected cells—appear to stick around longer. In a new British study, which hasn’t yet been peer-reviewed, researchers measured T-cell activity in 100 U.K. health-care workers six months after they had tested positive for Covid-19. None had been hospitalized; about half had been asymptomatic.  All 100 participants showed “robust T-cell responses,” even those whose antibody count had dropped to an undetectable level.

The scale of T-cell response was 50 percent higher in people who had shown symptoms of the disease. That could mean people with asymptomatic cases may not retain as much cellular protection or may simply be better equipped to fight off the virus without a big immune response. The authors emphasize that the study proves only that the T-cells are active, not that they confer immunity. “Cellular immunity,” lead author Shamez Ladhani told CNBC.com, “is a complex but potentially very significant piece of the Covid-19 puzzle.” 
--The Week
How to fix our election chaos

The ongoing chaos over the presidential election results is entirely unnecessary. The U.S. is unique among advanced nations in having each state hold its own election, with its own rules and method of counting and results coming out in dribs and drabs over days. Those varying rules led to wild fluctuations as results were updated [recently], with the “mirage” that President Trump was leading in several states; in reality, votes in Biden’s favor had already been cast and were simply waiting to be counted. The delays in counting mail-in ballots and votes from big cities allowed partisan conspiracy theorists to claim sinister forces were at work. This system needs to be brought into the 21st Century before the next presidential election.

In Canada, elections are conducted by a single, nonpartisan federal agency, with the same rules for every part of the country. Rather than release interim vote totals at random, a centralized system tabulates all the votes and only then announces who won. Our current hodgepodge system produces excruciating delays and confusion. For the sake of our psychological health and confidence in democracy, we desperately need reform.
--Law professor Stephen Vladeck, New York Times 

The polls: Why so wrong again?

#####Professional pollsters blew it again. Just as they did in 2016, the polls underestimated President Trump’s strength as they predicted an easy victory for the Democratic nominee. Some polls had Joe Biden ahead by as many as 10 percentage points nationally, while the final margin will probably be a much slimmer 4 to 5 points. On the state level, pollsters predicted a 3- to 6-point Biden victory in Florida and a toss-up in Texas. Trump took Florida by 3-plus and Texas by almost 6. Ohio and Iowa were rated toss-ups, and Trump took both by 8-plus. 
--Matthew Rozsa, Salon.com

#####Down ballot, the polling performance appeared even worse. Pollsters predicted a blue wave would carry Democrats to a gain of about a dozen House seats and three to six in the Senate. Instead, Democrats lost House seats and could end up gaining only one in the Senate, depending on the results of two runoff elections in Georgia. Maine Sen. Susan Collins—deemed a potential loser—wound up winning by almost 9 points. 

--Steven Shepard, Politico.com 

#####What a black eye for polling. After 2016, pollsters tried to capture more potential Trump supporters by weighting for rural voters and those without four-year college degrees. But the correction may have been insufficient and failed to compensate for the deep distrust many conservatives feel for the mainstream media and polling outlets. They may be hanging up on polling calls in disproportionate numbers.
--Aaron Zitner, Wall Street Journal 

#####Polling clearly has some serious challenges. Statistical analysis works only if you can reach representative samples of different voting groups. But if some groups decline to participate, polls will be consistently wrong.
--Nate Cohn, New York Times 

#####If that’s the case, polling faces an existential threat. But the damage won’t be limited to polling companies, which do much more than predict the outcome of elections. Much of American democracy depends on being able to understand what our fellow citizens think on such issues as gun control, religious belief, racial justice, climate change and abortion. Officials use polls to help decide policy and candidates to craft campaigns. As Americans sort themselves into ideological bubbles, the need for reliable polling is arguably greater now than ever before. If everyone decides polls can’t be trusted anymore, we’re all flying blind.

--David Graham, TheAtlantic​.com 
Look-alike couples

Back in 1987, scientists at the University of Michigan put out a study that suggested the faces of people in long-term relationships look more alike over time. Now a team at Stanford University has probed that theory and found it to be bunk, reports The Times (U.K.). By trawling through wedding anniversary announcements in newspapers, the Stanford team created a bank of pictures of more than 500 couples at different stages in their marriage.

They then used a computer algorithm and surveys with volunteers to determine whether the couples’ faces appeared to grow more alike. The researchers found clear evidence that when couples first get together, they are more likely to resemble one another than random pairs of people. But there was no evidence for them growing more alike over time. Lead author Pin Pin Tea-makorn says there are several explanations why people initially gravitate toward their look-alikes. “The first is that people tend to like things they are familiar with,” she says. “Another is that, in general, organisms tend to select a mate who is pretty similar to make sure they are not mating with a different species.”
--The Week 

Is the 4-percent rule too strict?

The man who invented the 4 percent rule has updated the numbers. In 1994, Bill Bengen, a financial adviser in Southern California, established the principle that “if you want to make sure your retirement savings last at least as long as you do, you should budget to spend no more than 4 percent of the balance in the first year—and then adjust the amount each year in line with inflation.”

But Bengen now says that this rule of thumb is too simplistic. He devised it as a hypothetical for someone retiring in October 1968, “the worst moment he could find in modern times,” just before a 14-year bear market and runaway inflation. Historically, however, the average safe withdrawal rate has turned out to be 7 percent. Bengen, a retiree himself, said he sticks to a withdrawal rate of 5 percent.
-- Brett Arends, MarketWatch.com 

Trump’s debt is a serious security risk

President Trump’s massive personal debt poses a significant counterintelligence risk. Recently published tax information shows he owes $421 million that he must pay off over the next few years, and by some estimates, his total debt may be as high $1 billion. As former national security officials, we’re keenly aware that our nation’s adversaries often use debt to compel Americans into betraying their country. Aldrich Ames, the infamous CIA spy who fed secrets to the Soviet Union, later admitted he did so because he was in debt and “needed funds to dig himself out of the hole.”

This is why Americans seeking access to classified information must submit to a deeply invasive review of their financial history, with two major exceptions: members of Congress and the president. Had Trump undergone the standard review, we have no doubt he would have been denied a clearance. Americans deserve to know whether the president owes money to Russians or other foreign actors. Is he making foreign-policy decisions with his faltering businesses in mind? Our country can no longer ignore the national security risks that Trump’s debt creates.
--Michael Morell and David Kris, Washington Post

How the modern dog came to be 

A major new study of prehistoric dog DNA has revealed how rapidly man’s best friend spread around the world following domestication and how the animals’ changed as they followed their two-legged masters. Researchers sequenced and analyzed DNA extracted from the remains of 27 ancient dogs unearthed in Europe, the Near East and Siberia. They found that by 11,000 years ago, just after the Ice Age and before the domestication of any other animal, there were already five different types of dogs with distinct genetic ancestries scattered across the world. The study suggests that domestication may have begun about 20,000 years ago, some 5,000 years earlier than previously thought.

Some of the researchers also believe the genomes show that all dogs evolved from an extinct and as yet unidentified form of wolf, though others say the evidence is not conclusive. The team made a pack of other findings, including that today’s dogs are much less genetically diverse than their ancient ancestors, presumably because breeders have prioritized powerful genes; that no new wolf DNA has entered dog genomes since wolves became dogs more than 15,000 years ago; and that dogs’ geographic spread didn’t always correlate with human migration. “In many cases humans would simply bring their dogs with them as they migrated and moved across the world,” co-author Anders Bergstrom, from London’s Francis Crick Institute, told CNN​.com. But “perhaps sometimes dogs were traded between human groups.”
--The Week 


No comments: