Lightnews — Scholar-powered news

David Hodgson

@dchodge.bsky.social

750 followers 960 following 48 posts

Researcher at Charité. Interested in mathematically modelling immunity. 🧘🍰🎧🖥️🏃🧪🏳️‍🌈 seroanalytics.org davidhodgson.me

Posts Media Videos Starter Packs

David Hodgson @dchodge.bsky.social · 2d

New blog post on Correlates of Protection! I try and give an overview of this very confusing concept: davidhodgson.me/post/sm3_cop1/

I think it's good ID modellers try and get a solid understanding of this as it's going become increasingly important for vaccine development over the next few years.

CORRELATES OF PROTECTION: FUMBLING THROUGH THE TERMINOLOGY | David Hodgson

Long ago, Edward Jenner made medical history by inoculating a young boy with cowpox and demonstrating protection against smallpox. Jenner had no idea why it worked, he just rolled with it, and thus in...

davidhodgson.me

David Hodgson @dchodge.bsky.social · 6d

Haha I actually switched to Claude this week so terse bluesky posts incoming.....

David Hodgson @dchodge.bsky.social · 6d

AHH cool! I'll have a play with this, doesn't seem active currently tho

David Hodgson @dchodge.bsky.social · 6d

Cheers Sam! I've not seen this have you got a link? They are fitting ODIN models with monty these days right?

David Hodgson @dchodge.bsky.social · 7d

Yeah it's actually great, converting the c++ to JavaScript is actually not too bad with a little help from AI !

David Hodgson @dchodge.bsky.social · 8d

Watch RJMC explore different model dimensions in real-time, use sample data or upload your own CSV.
No installation, just open and experiment. Great for teaching/learning Bayesian model selection!

Package/vignette: dchodge.github.io/rjmc/article...
#statistics #bayesian #MCMC #datascience

Example 1: Mixture model

dchodge.github.io

David Hodgson @dchodge.bsky.social · 8d

🚀 New tool: Reversible Jump MCMC running in your browser!
Built an interactive widget for fitting mixture distributions when you don't know how many components you need.

Check it out: dchodge.github.io/rjmc-widget-...

Dynamic Mixture Model Analysis

dchodge.github.io

David Hodgson @dchodge.bsky.social · 10d

🔬 New to serological data? You’re not alone

When I first saw spreadsheets full of columns labelled ELISA_OD, PRNT50, HI_titre, and PVNT_ID50, I had no idea what they really meant.

That confusion inspired me to write a new blog post, “A Dummy’s Guide to Serological Assays”

👉 tinyurl.com/586dsy77

This link will take you to a page that’s not on LinkedIn

David Hodgson @dchodge.bsky.social · 12d

Sure, wanna drop me an email to sort out deets?

David Hodgson @dchodge.bsky.social · 13d

"2.1. Overview of inference framework" in the methods gives an overview. But basically if you infer an infection you also need to infer an infection time (an extra parameter), no infection then infection time isn't in the framework anymore. Hence need to jump between different dimensions

David Hodgson @dchodge.bsky.social · 15d

Thanks to everyone who worked on this: @jameshay.bsky.social, Sheikh Jarju, Dawda Jobe, Rhys Wenlock, @adamjkucharski.bsky.social, and @thushan-desilva.bsky.social!

David Hodgson @dchodge.bsky.social · 15d

It uses reversible jump MCMC to infer missed infections, to help understanding I made a little widget to show you how the fitting process works for simulated data: seroanalytics.org/serojump-widget

Interactive widget for serojump

seroanalytics.org

David Hodgson @dchodge.bsky.social · 15d

✨ What we did:
- Made a Bayesian model to infer who was infected, when, and how their antibody levels changed
- Validated on both simulations and real-world SARS-CoV-2 data from The Gambia.
- Showed that serojump detects more infections (including sub-threshold ones) and provides richer insights

David Hodgson @dchodge.bsky.social · 15d

🚨 New paper out in PLOS Computational Biology! 🚨

We're excited to share our new paper, serojump, a new probabilistic framework and R package for inferring infections and antibody kinetics from longitudinal serological data.

📄 Full paper: tinyurl.com/re7du3t2
R package: seroanalytics.org/serojump

A serological inference package using reversible jump mcmc

The `serojump` package provides tools for fitting serological models to antibody kinetics data using reversible-jump Markov Chain Monte Carlo (RJ-MCMC). It enables researchers to model the dynamics of...

seroanalytics.org

David Hodgson @dchodge.bsky.social · 15d

Thanks to every who worked on this! @jameshay.bsky.social, Sheikh Jarju, Dawda Jobe, Rhys Wnelock, @adamjkucharski.bsky.social and @thushan-desilva.bsky.social

David Hodgson @dchodge.bsky.social · 15d

serojump was designed to be a flexible and pathogen-agnostic solution that can be applied to a wide range of pathogens.

Heres an interactive widget to help with understanding of them reversible jump mcmc methods: lnkd.in/eWGJ39PG

This link will take you to a page that’s not on LinkedIn

David Hodgson @dchodge.bsky.social · 15d

What we did:
- Made a Bayesian model to infer who was infected, when, and how antibody levels changed over time.
- Validated on both simulations and real-world SARS-CoV-2 data from The Gambia.
- Showed that serojump detects more infections (including sub-threshold ones) and provides richer insights

David Hodgson @dchodge.bsky.social · 18d

Key features:
- WebAssembly-powered performance (10-50x faster than JS)
- Adaptive MCMC for Bayesian inference
- Vaccine intervention analysis with waning immunity
- Real-time convergence diagnostics
- Export data and plots for further analysis

David Hodgson @dchodge.bsky.social · 18d

Just launched an interactive Bayesian epidemic modelling platform that runs entirely in your browser!

No downloads, no installations, no expensive software licenses. Just open the link and start modelling disease dynamics with real-time parameter estimation.

>> widget-bayesian-sir.davidhodgson.me

Interactive Bayesian Epidemic Modelling

widget-bayesian-sir.davidhodgson.me

David Hodgson @dchodge.bsky.social · 21d

Thanks Sam! All code for the website and blog posts is here github.com/dchodge/David.io !

GitHub - dchodge/david.io: davidhodgson.me

davidhodgson.me. Contribute to dchodge/david.io development by creating an account on GitHub.

David Hodgson @dchodge.bsky.social · 21d

What does it really mean to be “infected”?

PCR? Symptoms? Antibodies? Being able to pass it on?

I’ve written a blog digging into why infection isn’t a simple on/off switch, and why the definition you choose matters for modelling, vaccines, & public health.

davidhodgson.me/post/sm1_wha...

WHAT DOES IT MEAN TO BE INFECTED? | David Hodgson

Defining infection is like reading tarot cards - the interpretation depends on who’s asking and what they’re looking for. You wake up with a sore throat. You take a lateral flow...

davidhodgson.me

David Hodgson @dchodge.bsky.social · 23d

🚨 New vignette: reversible-jump MCMC for epidemic change-points

Used RJMCMC to infer piecewise attack rates in an SEIR model — the sampler figures out how many change-points there are and where they happen.

Code + walkthrough 👉 dchodge.github.io/rjmc/article...

#R #Bayesian #Epidemiology #RJMCMC

Example 2: SEIR change-point attack rate

dchodge.github.io

Reposted by David Hodgson

Adam Kucharski @adamjkucharski.bsky.social · Apr 9

We've got a new pre-print (spearheaded by @dchodge.bsky.social) on how to reconstruct unobserved antibody kinetics and infections - and use these to estimate correlates of protection: www.medrxiv.org/content/10.1...

And here's the accompanying {serojump} R package: seroanalytics.org/serojump/

Reposted by David Hodgson

Adam Kucharski @adamjkucharski.bsky.social · Mar 31

Why did neutralising antibody responses vary between the Oxford-AZ and Pfizer-BioNTech SARS-CoV-2 vaccines? And how did changing the duration between first and second dose affect the underlying biological processes?

Our new paper, spearheaded by @dchodge.bsky.social had a look.... 1/

Reposted by David Hodgson

Alex Hill @alexlizhill.com · Mar 6

@dchodge.bsky.social and I developed an app using WebR and React, to see how this stack compares to RShiny. Looks quite promising I think!

The app wraps the SeroSim R package by @jameshay.bsky.social for simulating serosurvey data:

serosim.seroanalytics.org

Web interface for serosurvey data simulation

serosim.seroanalytics.org