This is now.

Sam Abbott@seabbs.bsky.social · yesterday

Tomorrows #epinowcast seminar.

#julia for infectious disease modelling workshop @lshtm.bsky.social in September community.epinowcast.org/t/topic/427

𝗥𝗲𝗴𝗶𝘀𝘁𝗲𝗿 𝗬𝗼𝘂𝗿 𝗜𝗻𝘁𝗲𝗿𝗲𝘀𝘁: 𝗠𝗼𝗱𝗲𝗹𝗹𝗶𝗻𝗴 𝗘𝗽𝗶𝗱𝗲𝗺𝗶𝗰𝘀 𝘄𝗶𝘁𝗵 𝗝𝘂𝗹𝗶𝗮 𝗪𝗼𝗿𝗸𝘀𝗵𝗼𝗽

We are hosting a free 3-day workshop on 𝟭𝟲𝘁𝗵-𝟭𝟴𝘁𝗵 𝗼𝗳 𝗦𝗲𝗽𝘁𝗲𝗺𝗯𝗲𝗿 𝟮𝟬𝟮𝟲 at 𝗟𝗦𝗛𝗧𝗠! Participants will learn how to use the Julia programming language to build deterministic models from first principles, st...

community.epinowcast.org

Some times when I read code for papers I find it hard to believe that people care so little for the craft of their work. I have to constantly remind myself that its lack of aesthetic sense doesn't always mean it is bad.

Tomorrows #epinowcast seminar.

Clara Brigitta@clarabrigitta.bsky.social · 2d ago

📢 Real-time Infectious Disease Modelling Seminar! Cathal Mills (Oxford) presents "Mechanism-led renewal equations", new methods addressing where classic renewal equations fall short. 🦠 #Epidemiology #IDModelling #PublicHealth

With various things in the news and maybe time passing, COVID discourse seems to be coming around again and the main thing I can share as a specialist is how different pandemic preparedness is from standard politics or almost all the other things in the Opinion columns. Primarily, a lot of >

Had a nice time reading arxiv.org/pdf/2607.058... which develops a new (I think) deconvolution approach and then takes the popular spline approach to Rt estimation.

I really think the solution has to be open source but not open contributor projects. A small barrier to being allowed to open a PR goes a long way.

Carlos Scheidegger@cscheid.net · 6d ago

So there's a thing happening on open-source repos: people who are otherwise uninterested in your project but want to obtain clout will file mediocre PRs against your repo anyway. So: community tarpits? Create a legit-looking repo with a bunch of open issues, wait for PRs, blacklist those accounts

Thought this claude generated documents my slide into LLM madness quite well

Sam Abbott@seabbs.bsky.social · 6d ago

Just gave a talk to the @cmmid-lshtm.bsky.social ECR group on how I llm (of course as is fitting written by an llm hoard) samabbott.co.uk/how-I-llm/

Just gave a talk to the @cmmid-lshtm.bsky.social ECR group on how I llm (of course as is fitting written by an llm hoard) samabbott.co.uk/how-I-llm/

In 2017, while working at Epicentre/Medecins Sans Frontieres I helped to develop the Urgepi program in southeast DRC. What started as measles surveillance program has matured into a prioritization scheme for outbreak prevention and response that serves 15m people (land area ~Spain).

Risk-targeted interventions for measles control- Lessons from an emergency response project in the Katanga Region, Democratic Republic of the Congo (2021-2023)

The Urgepi project is a measles emergency initiative implemented by Medecins Sans Frontieres Operational Centre Paris (MSF-OCP) in the Katanga region of the Democratic Republic of Congo, a setting cha...

medrxiv.org

What is an 'everyday task' and what is a 'complex problem' in bioinformatics software engineering? Feels more important use the right model now they're not so heavily subsidised! I did a small comparison on an optimisation task: johnlees.me/posts/ska2-c... Small/cheap model wasn't good enough

Had a fine time today teaching forecasting hubs with Nick Reich at #SISMID (good org from @nataliexdean.bsky.social and co). Let seabbs-bot wild on the local hub playground to compete with the students with some interesting results: github.com/seabbs-bot/s...

GitHub - seabbs-bot/sismid-ili-turing: Hierarchical Bayesian (Turing.jl) ILI% forecasting for the SISMID reichlab sandbox hub: partially-pooled seasonal + AR/VAR dynamics with a joint backfill model, ...

Hierarchical Bayesian (Turing.jl) ILI% forecasting for the SISMID reichlab sandbox hub: partially-pooled seasonal + AR/VAR dynamics with a joint backfill model, searched and selected by WIS. - seab...

github.com

Yesterday marked two months since the DRC declared the ongoing Ebola outbreak, @drtedros.who.int says in presser. "It's now the third largest Ebola outbreak on record, and in the past months, it has expanded faster than any previous outbreak."

Looking forward to teaching the nowcasting half of our nowcasting and forecast course at SISMID starting today! Some nice (I hope) new content on modelling workflows vibing on the Bayesian workflow book from @avehtari.bsky.social et al. and new session on complex reporting processes.

New #epinowcast release including support for ARIMA latent processes. New contribution from Nick Davies fixing an issue they found whilst doing some response work github.com/epinowcast/e...

1/ For those attending @ecmtb2026.org, I will be presenting three talks during the conference. #IDsky #EpiSky 🧪 🔗 ECMTB 2026: ecmtb2026.org 👇 See thread for further talk details and associated papers.

European Conference on Mathematical & Theoretical Biology 2026 | ECMTB 2026

Join us in Graz, Austria, from July 13–17, 2026, for the 14th European Conference on Mathematical and Theoretical Biology (ECMTB'26)!

ecmtb2026.org

Really enjoyed giving a short workshop on what is a forecast and is it any good at #iddconf. I lasted attended (I think) 8 years ago as Phd student and loved it so looking forward to the talks and community. Our workshop material is here: nfidd.github.io/iddconf2026/

PhD position (Melbourne, Australia) Mechanistic modelling of the epidemic dynamics of vector-borne zoonotic diseases in Australia with Oliver Eales, Freya Shearer, Marya Potorek at University of Melbourne More details: http://iddjobs.org/jobs/2561

epinowcast seminar now on infectious disease modelling crossed over with survival analysis.

Clara Brigitta@clarabrigitta.bsky.social · last mo.

🗓️ When: July 2nd, 3pm BST / 4pm CEST 🔗 Link to join: lshtm.zoom.us/j/8290218109

Looking forward to this tomorrow at 3pm uk time. The workshop was great.

Clara Brigitta@clarabrigitta.bsky.social · last mo.

Are you interested in survival analysis or infectious disease modelling? Ever wondered how the two could contribute to one another? 🤔 Well come along to the next Epinowcast seminar tomorrow!

Great to see {epidist} and so @paulbuerkner.com {brms} and @mc-stan.org getting used as part of the reporting of the clinical characteristics of the Bundibugyo Virus outbreak in the DRC.

New paper: We recast automated scientific model discovery with LLMs as Bayesian inference! LLMs write code and carry domain knowledge, great for proposing models. The key idea: discovery is inference, not just generation. What distribution of models explains the data? 🧵 arxiv.org/abs/2602.18266

and yet....

ShitMyReviewersSay@shitmyreviewerssay.bsky.social · last mo.

"I doubt that it’s worth publishing a handful of graphs that few would find surprising and that anyone else could generate in a day off."

A new paper by @dominiquemakowski.bsky.social, @mattansb.msbstats.info, me and colleagues just out! We show how to choose informative priors in Bayesian regression models using a systematic simulation study and a practical step-by-step tutorial in #Rstats and #Stan! doi.org/10.3389/fpsy... >

Nice to have a new EpiNow2 release with lots of work from @sbfnk and others. epiforecasts.io/EpiNow2/news...

Postdoc (Nottingham, UK) Develop the next generation of tools for fitting stochastic epidemic models to outbreak data. with Theodore Kypraios, Philip O’Neill at University of Nottingham More details: http://iddjobs.org/jobs/2557

Enjoying the survival analysis x infectious disease modelling workshop in Leiden! Gave a talk today: samabbott.co.uk/how-to-seria... Used it as a vehicle to throw in some asks - answers please!

A few weeks ago @sbfnk.bsky.social was doing some re-estimation work for www.nejm.org/doi/10.1056/... (lovely dataset and paper) We just landed a real-time version of the model (i.e. that you could use in real-time and not (ideally) get biased estimates.

“Super-Spreaders” and Person-to-Person Transmission of Andes Virus in Argentina | NEJM

From November 2018 through February 2019, person-to-person transmission of Andes virus (ANDV) hantavirus pulmonary syndrome occurred in Chubut Province, Argentina, and resulted in 34 confirmed infe...

nejm.org

Found my first vibed up port of some of my work today into another language stack (without the statement that is what it is of course but looks fairly clear). Very interesting to think about what this means for the future.

This is happening now

Clara Brigitta@clarabrigitta.bsky.social · 2mo ago

Happening today! 3pm BST / 4pm CEST Come hear some lightning talks on modelling/modelling-adjacent work relevant to the current Ebola outbreak, and stay for the discussion! We'll leave plenty of time for it 🌟 community.epinowcast.org/t/community-... #Ebola #BVD #PublicHealth #InfectiousDiseases