Long weekend; just catching up

Saturday and Sunday, the Apollo Chorus sang Verdi's "Requiem" three times in its entirety (one dress rehearsal, two performances), not including going back over specific passages before Sunday's performance to clean up some bits. So I'm a little tired.

Here are some of the things I haven't had time to read yet:

• I always read Andrew Sullivan's weekly column but I haven't had a chance yet.
• Democratic candidate Conor Lamb might win in a heavily-Republican district in Pennsylvania. (Disclosure: I have contributed to Lamb's campaign.)
• The Telegraph UK reports that there is now a genetic test for intelligence.
• Jeet Heer returns to political reporting, suggesting that the President's ideas aren't as bad as his execution of them.
• Now that we're back on Daylight Saving Time in North America, The Daily WTF reminds us how much fun it is to debug software implementations of time zones.
• Microsoft's Raymond Chen started a series of blog posts on advanced Git management.
• Cranky Flier says American's most recent changes to its scheduling practices are good for customers.
• Closer to home, the city of Chicago and the Chicago Transit Authority have started construction aimed to fix one of the biggest bottlenecks on the El.
• Crain's Chicago Business reports that, for the first time in its existence, Groupon has had a profitable quarter.
• Chicago History Today has side-by-side photos of Dearborn and Jackson from 1936 and 2017.
• And finally, NASA has launched two new weather satellites that will significantly advance weather forecasting.

Other stuff is going on, which I'll report when I have confirmation.

The Daily Telegraph's headline is highly misleading. It is a gross simplification of a very complex subject. The scientific research which is reported in the paper is an example of a genome-wide association study (GWAS). These kinds of studies are used to try to identify genetic links to diseases such as heart disease which do not have single-gene causes (unlike, say, cystic fibrosis or haemophilia). They have become popular in the past few years because genome sequencing technology now allows an entire human genome to be sequenced quickly and cheaply, and then compared to a "reference" genome to identify single-nucleotide differences, much as a software developer might use the 'diff' tool to find the differences between two versions of the same source code file. A typical GWAS takes DNA samples from a large number of people who suffer from a particular medical condition, plus samples from a control group of people without the condition. All of the samples are sequenced and compared to a reference genome. All of the single-nucleotide differences are identified, and statistical methods are used to sift out specific differences which appear predominantly in the patient group but not in the control group. These differences may not actually occur within a gene, but rather, in the vast stretches of DNA which lie between genes. The study mentioned in the Telegraph article appears to have used a standardised verbal, numerical and reasoning test as a proxy for "intelligence". This kind of test was once used in Britain on 10-year-olds to decide what type of school they should attend from age 11: if you passed the test, you went to a grammar school where the curriculum was oriented towards sending you to university at age 18; if you failed, you went instead to a school with a more vocational curriculum with the expectation that you would leave at age 16 and start work in a blue-collar job. This practice was abandoned forty years ago. Anyway, the outcome of the study was that from a cohort of around 70,000 subjects, they identified over 500 genes where single-nucleotide changes seemed to be correlated with the subjects' performance in the test. So what they have done, in fact, is found that the ability to get high scores on a very specific type of cognitive function test may *possibly* be linked with *hundreds* of different genes. Each subject will likely have had a different combination of variants of those genes. A GWAS cannot drill down to the level of an individual, sine it relies on statistical analysis of large cohorts.