John Belmont

Had a fantastic time talking with @lucystats.bsky.social and @epiellie.bsky.social about data science communication, causal work in academia vs industry, and more! (wherein I also soft commit to ~3 WIP blog posts 😬)

One of my all time favorite podcasts — downside is I get one fewer net new episodes

Do they really not know about ctrlF ?

The Epstein Cover-Up at the FBI Inside the chaotic review process of the Epstein and Maxwell files at the Bureau

The amazing thing about brave people speaking out is that it inspires others to do the same by creating a permission structure. Since I published this article 4 hours ago, I've heard from multiple sources confirming what's in it. www.muellershewrote.com/p/the-epstei...

Big Bend looks safe

And when it's all said and done, he will probably go right back to being Johns Hopkins Professor of Medicine Marty Makary.

DiffMethylTools: a toolbox of the detection, annotation and visualization of differential DNA methylation.#DNAmethylation #DifferentialMethylationAnalysis #Bioinformatics #Genomics @biorxiv-bioinfo.bsky.social 🧬 🖥️
www.biorxiv.org/content/10.1...

At this point, I might as well --
Here's an infographic showing different ways to include age as a predictor. The top shows two extremes, just as a plain old numerical predictor (imposes linear trajectory) vs. categorical predictor (imposes nothing whatsoever). And then three solutions in between!

Five more today. It’s warm and muggy in the early morning.

Ted Cruz ensured Trump spending bill slashed weather forecasting funding Senator on vacation abroad while Texas was hit by deadly floods, a disaster worsened by forecasting cuts, critics say

Sen Cruz inserted language into the Republicans’ “big beautiful” bill to eliminate funding for improvements in weather forecasting and dissemination of forecasts & warning to the public.
www.theguardian.com/us-news/2025...

A Clinical Genetics and Genomics review course, if that is how you want to spend your Saturday morning (works for me!).

A day that begins with a night bloomer is promising 🌱

Texas floods death count surpasses 100 as recovery efforts continue Camp Mystic says it’s grieving 27 campers and counselors. At least two dozen people were reported missing across the region Monday, though it is not clear how many people in total are still unaccounte...

The number of people killed in devastating Central Texas floods surpassed 100 on Monday as emergency officials spent another day searching for those who remain missing — and as clouds threatened to bring more rain and flooding to the region.

This is what I was talking about in my opinion piece for MSNBC over the weekend. By launching these detention partnerships with other countries—the administration claims to currently be in negotiations with dozens of countries—the US is spurring the creation of a global concentration camp network.

link 📈🤖
Efficient Estimation of Causal Effects Under Two-Phase Sampling with Error-Prone Outcome and Treatment Measurements (Barnatchez, Josey, Hejazi et al) Measurement error is a common challenge for causal inference studies using electronic health record (EHR) data, where clinical outcomes and

link 📈🤖
Degree of Interference: A General Framework For Causal Inference Under Interference (Ohnishi, Karmakar, Sabbaghi) One core assumption typically adopted for valid causal inference is that of no interference between experimental units, i.e., the outcome of an experimental unit is unaffected

Grant applications

It could be like designing an antibody to reduce circulating cystatin C levels in an attempt to improve kidney function...

Fun with Positron | Andrew Heiss Combine the best of RStudio and Visual Studio Code in Posit’s new Positron IDE

I’ve been using Positron for the past ~9 months and love it. I was already using VS code for everything except R—RStudio’s plot panel and environment panel couldn’t be replicated there. But Positron does that!

See here for more too www.andrewheiss.com/blog/2024/07...

I just have to say, as a historian of the Reconstruction-era Klan, that masked paramilitaries are very bad and masked vigilantes are even worse and there are causal links between the two and none of this is good for democracy.

lol, first time a co-author added a reference to Lundberg et al. (2021) before I had a chance to do so. Already an all-time classic. The opening of the abstract also remains an absolute banger.

journals.sagepub.com/doi/10.1177/...

After two weeks, I'm finally done!

In this post, I explain different approaches for solving linear regression in R: directly, using QR, singular value and Cholesky decompositions, and do some benchmarking for comparison with in-built approaches.

thomvolker.github.io/blog/2506_re...

YouTube video by SlyATFamilyStoneVEVO Sly & The Family Stone - Stand! (Official Audio)

Loved Sly and the Family Stone.

Stand fits the current moment youtu.be/Q49vjFN6Fsw?...

The most bizarre thing about the COVID School Closure Discourse is how commentators take it for granted that closing schools had no effect on transmission or deaths but that's not at all what the evidence says.

As always, Dr. Murray is insightful about the details and why big conclusions often hang on them.

The MAHA document is particularly disgraceful in the way that it uses legit concerns about children's health to advance a slew of half-baked ideas. Did they consult even one pediatrician?

E is for Epi | Ellie Murray, ScD | Substack Making sense of the science of public health. Click to read E is for Epi, by Ellie Murray, ScD, a Substack publication with thousands of subscribers.

Recommending the new E is for Epi epiellie.substack.com where Dr. Murray is currently reviewing the MAHA children's health report. This series dissects the details of a document that is supposed to be a centerpiece of RFKjr's health agenda.

Here Are the Nearly 2,500 Medical Research Grants Canceled or Delayed by Trump (Gift Article) Some cuts have been starkly visible, but the country’s medical grant-making machinery has also radically transformed outside the public eye.

The destruction of the medical research enterprise in the U.S. will go down in history as one of the most damaging choices a government has made in setting its population backwards
www.nytimes.com/interactive/...

We need to see what’s left of NIH once this regime is booted out.

If there are still some grants to be had in the meantime, give some consideration to survival of your field.

I had an explicit career goal to either use a comercial role or simply age out of the NIH grant system. I want younger scientists to preferentially use those resources.

Now with DOGE/RFK catastrophe I urge late career scientists to creatively contemplate new roles for themselves.

Or do you like this one