I think news stories like this have been so frequent over the last six years that it's now "common knowledge" that covid ages your internal organs. But, it's simply not true. 1/

An example of how impacts of interventions can differ at individual and population levels, and are conditional on other events. Given a car hits you, the helmet helps. But mandating helmets could lower the # of bikes on the road & thus might increase the chance of getting hit in the first place.

New Preprint! We look at HIV transmission outside of and within stable partnerships. It turns out that the pair model can have many useful results derived analytically in closed form, and the general numerical methodology should be useful in other contexts. arxiv.org/abs/2602.04638

Trump’s $12B bet — Project Vault — on which minerals will be of future technological importance aims to provide certainty to industry. But does it risk locking us in to the status quo, when materials discovery is the one field I can say will genuinely be disrupted by quantum computing and AI… 🤔

Landing in Dunedin Aotearoa in about 90 minutes for the start of a trip that represents the culmination of a dream I’ve had since I was ten years old.

Petition to replace electric heaters with in-house servers that can, in cold weather, be put at the disposal of tech giants for reasonable rates. Free heating and less power demand - win win!

New pre-pre-print:

Pleased to say that Carl Bergstrom (@carlbergstrom.com) will be speaking at the University of Canterbury on 13th Feb! He is hosted by @tepunahamatatini.bsky.social come see him at noon in Jack Erskine 441!

Pretty extraordinary to think about how much cardiovascular researchers figured out, and how that turned into public health campaigns, medicines, surgeries, and emergency care that changed millions of people's lives. ourworldindata.org/cardiovascul...

Just a few months ago, a child in the US developed the most feared complication of measles (subacute sclerosing panencephalitis) from an infection they had years ago when they were too young to be vaccinated. I hate to think about what is being unleashed by this virus as a future tragedy.

Very happy to have had the opportunity to contribute to this work. Thanks to Robin Thompson for leading, the amazing team of co-authors and the INI for hosting us.

Missed this in the end of year rush - Vincent Lomas's first paper from his PhD looks at modelling Covid-19 transmission dynamics in different ethnicity groups in Aotearoa New Zealand

Jew in Amsterdam here. Please use Anne Frank as a comparison point to explain how state persecution of scapegoats impacts children, that is the point of her example, thanks

And some wonder how a hard-to-transmit camel cold virus keeps popping up in humans in the Arabian Peninsula. A picture tells a thousand words.

Very happy to have had the opportunity to contribute to this work. Thanks to Robin Thompson for leading, the amazing team of co-authors and the INI for hosting us.

Missed this in the end of year rush - Vincent Lomas's first paper from his PhD looks at modelling Covid-19 transmission dynamics in different ethnicity groups in Aotearoa New Zealand

New paper! "Infectious disease outbreak controllability: biological, social and public health factors" - we provide a modelling-informed discussion of what 'controllability' means for an infectious disease outbreak. royalsocietypublishing.org/rspb/article...

Evergreen reminder that this was something she came up with, not something she was advised to do. In fact she was pretty specifically advised it was a terrible idea (it was) but she went ahead with it anyway because something something Not Real Science

These are my favorite kind of modeling studies. A lot of people want models to precisely predict the future, but the best models help give us a framework for understanding what we are observing in real time, which in turn can inform expectations about potential future outcomes.

In a new preprint, we combine modeling with school-level vaccine data to contextualize the risk of 'breakthrough infections' and impacts on the ongoing US measles outbreak. #idsky #episky www.medrxiv.org/content/10.6...

3. And to do that, you want to *understand* and *justify* every step along the way (at least until you reach one of Latour's Black Boxes) because your end product is a theory, model, or explanation. You can't readily just hit RUN and see if it sorta works.

1. The thing about science that these jokers don't understand is that science cannot be vibe-coded. Whatever its flaws, the point with vibe coding is that you're trying to quickly make something that sorta works, where you can immediately sorta see if it sorta works and then sorta use it.

Te Pūnaha Matatini is bringing acclaimed evolutionary biologist and science system researcher @carlbergstrom.com to New Zealand in February 2026 for a series of public engagements, including a keynote address at the Capital City Complex Systems Symposium → www.tepunahamatatini.ac.nz/posts/carl-b...

Hey Aotearoa/New Zealand academics: Here's something good you can do today! Only 1 of 8 universities in AoNZ has left X. I think it should be 8 of 8. (So far 10 of 37 Australian public unis left X - you're behind 😬) Full grade report for AoNZ unis, and what you can do to help, follows.

Regardless of the usefulness of "AI" based tools for progressing science, there can't be any such thing as a "AI co-scientist" because scientist is a term that describes people, not things, and we absolutely must not lose sight of that. It's not just semantic.

Really excited our paper about how H5N1 rapidly adapted to cattle (and how these adaptations also increased its ability to infect cells from the human respiratory tract) is now published! www.nature.com/articles/s41...

OUP want $8500 to publish a 400 word letter to the editor 🤣

Dont miss the ESPIDAM summer program in June 2026 in Stockholm, covering many key concepts for ID modelling: stochastic models, AI for ID control, nowcasting and forecasting, phylodynamics, data analysis, network models, within-host models, health economics www.statistics.su.se/english/divi...

I don't think most Americans know this: "[Denmark] lost more soldiers per capita than the U.S. did in Afghanistan". (in fact, the most of any). Why? Because, Denmark as a loyal ally, came to the aid of the U.S. Now, the U.S. wants to invade Danish territory. www.theatlantic.com/national-sec...

Here in AoNZ, we will get year to December data mid February, but age standardised mortality for year to Septembers looks like 2025 is going to be about 4% lower than 2024 (and through improvements over time, next year will be the year mortality improves on the low we saw in 2020)

The CMI has released its analysis of 2025 mortality in England & Wales. Death rates for 2025 were nearly 2% lower than 2024, the previous record low. Death rates improved (i.e. fell) for both sexes and in all age groups between 2024 and 2025.

That we're in a place, at a time, when someone has to write this headline, this article — it's almost unfathomable. Please do read @krutikakuppalli.bsky.social's take. www.statnews.com/2026/01/12/m...

It was not a super flu year and why I think government and NHS messaging was wrong and counterproductive - my latest post christinapagel.substack.com/p/it-was-not...

I have no interest in trying to dunk on anyone but there is absolutely no evidence for this at all. SARS-CoV-2 has not been demonstrated to make you more susceptible to other infections by impairing the function of your immune system.

Fiordland 😍

Status update: still on holiday Wakatipu edition

Extended influenza seasons in Australia and New Zealand in 2025 due to the emergence of influenza A(H3N2) subclade K viruses www.eurosurveillance.org/content/10.2...

This influence campaign is coming to Ireland. Our leaders in higher education must get out in front of it. The evidence is clear. Pursuit of diverse knowledge for its own sake helps drive prosperity. Now go tell people!

Interesting perspective on the recently amended International Health Regulations in relation to NZ's response to Covid. One size doesn't fit all & the IHR have enough flexibility for countries to determine their own response while still fulfilling intl obligations link.springer.com/article/10.1...

In Ireland we probably averted about 500 newborn RSV hospitalizations once nirsevimab was introduced last year (~75% risk reduction). Uptake was > 80%, and that's what's threatened by these cynical safety concerns from the antivaxxers in the US government. www.medrxiv.org/content/10.1...

Foraging ecology drives viral community structure in New Zealand's aquatic birds www.biorxiv.org/content/10.6...

So, some people here are concerned with the sentence in this document that, "it will not be possible to halt the spread of a new pandemic virus, and it would be a waste of public health resources and capacity to attempt to do so" - what does it this mean? 🧵/1 www.england.nhs.uk/long-read/fr...

Pope publicly accused of Catholicism

Their rationale, surprise, is economic cost. Maybe look at what the economic cost of *not* reducing emissions is going to be? Inaction on climate is buy-now, pay-later - and like all credit it costs you more in the end.

Shipping and aviation emissions excluded from NZ’s climate targets – Expert Reaction

Australia's wild flu season continues: Flu at epidemic levels in New South Wales heading into the first week of summer, sending more people to ED weekly than covid did at any time in 2025. This is driven by the new subclade-K variant of A/H3N2. 1/

This is interesting from OpenSAFELY. A thorough analysis of associations between COVID and gastrointestinal diseases. I suspect there are similar with other viruses called "respiratory" like RSV but it's harder to do the studies.

A new postdoc position just opened in our group at NAU! We are looking for someone to lead our current forecasting activities and the development of our next generation AI/ML models! Don't hesitate to reach out to me with any questions! careers.nau.edu/jobs/postdoc...

cool confounding and adjustment example by @jofrhwld.bsky.social below. To estimate A –direct–> Y, must adjust for B (and C or D). If you adjust for C, it partly opens collider D, so non-causal path D <– B –> Y is opened. Easy to forget about descendants partly opening parents. Sneaky colliders.

Auckland Uni has responded to my OIA request re consultancy spending in 2024-2025 $600k to PWC and $500k to Nous Group for "strategic design, organisational change, transformation services" etc This is roughly what our whole faculty of science spent on internally funded postdocs in the same years