Elan Bechor

Complex traits and preimplantation genetic testing. Formerly @OrchidInc @GenomPrediction. PhD @UCBerkeley.

[Elan Bechor @elanbechor]

@harryh FWIW, a quick waterfall plot suggests that enrollment decline is most of the story.

[Elan Bechor @elanbechor]

Because of the Nikole Hannah-Jones story from yesterday, people noticed that New York spends more than $40,000 per student.

I was curious about the cost disease behind this.
Answer: for decades spending has grown +2.2% per year above inflation for multiple reasons. (1/4)

[Elan Bechor @elanbechor]

@harryh Is there existing data that lets us separate those effects? I don't know enough to say.

[Elan Bechor @elanbechor]

Equally important has been +82% real growth in healthcare spending, which is fully subsidized, i.e. $0 employee premiums for individual and family plans (4/4)

[Elan Bechor @elanbechor]

Pension costs were growing very rapidly then started declining in 2019 due to changes in actuarial methods strong post-Covid returns (3/4)

[Elan Bechor @elanbechor]

Because of the Nikole Hannah-Jones story from yesterday, people noticed that New York spends more than $40,000 per student.

I was curious about the cost disease behind this.
Answer: for decades spending has grown +2.2% per year above inflation for multiple reasons. (1/4)

[Jack @tracewoodgrains]

This is the per-pupil funding for the elementary school Nikole Hannah-Jones sent her daughter to.

If you subscribe to the funding theory of education, how much would have been enough? It's already crossed the average per capita annual income threshold. Maybe $100k? $200k?

[Elan Bechor @elanbechor]

@CJHandmer @TerraformIndies Congratulations! Amazing and inspiring work.

[Elan Bechor @elanbechor]

@GoodPyre No, that is just what a local ancestry classifier gives using modern populations. Using the same method, you might conclude French and German people are mostly British and Scandinavian.

[Elan Bechor @elanbechor]

Nick, with all due respect: this is an incredibly amateurish understanding of what PCA tells you.

The "cline" in a two dimensional PCA plot is heavily dependent on the populations you put in the panel.

[Nassim Nicholas Taleb @nntaleb]

In the process of putting a dent in the theory that "Ashkenazis have some Levantine origin".
The claim: "The AJ cluster sits on the Europe–Levant cline betw S. Italians & Levantines on PC1/PC2."

No, if the "cline" includes the Caucasus, then it sits BEYOND Georgia/Armenia, ar. the BLACK SEAN.

[Elan Bechor @elanbechor]

@USA_casserole @gen0m1cs Did you read my second post?

[Elan Bechor @elanbechor]

If you want to actually understand the genetic origins of Ashkenazi Jews:

1. Xue et al showing the Levantine origin of the non-European component
https://t.co/NoE4mE5SlH

2. Waldman et al. 2022
https://t.co/NUS58cQVgQ
but don't take "65% Southern Italian" too literally

[Elan Bechor @elanbechor]

Two other thoughts:

First, there is very little information here about their results aside from the headline number, not even how they measure within-family R^2.

Second, it's very petty to not cite the cogPGT paper results from @_twolfram et al in the benchmark table.

[Elan Bechor @elanbechor]

Interesting whitepaper from @nucleusgenomics claiming an IQ PGS that matches @Herasight's cogPGT.

One thing that jumps out at me: they validated on the UK Biobank, which has been down for about 5 months now and is still closed.

[Kian Sadeghi @KianSadeghi5]

@sebuzdugan @nucleusgenomics all our scores are validated within sibling

Here’s the white paper for Vitruvian specifically:

https://t.co/74zZmyfUuC