What makes bird flu spill over to humans? 4y funded PhD in the genomics and molecular virology behind zoonosis of avian influenza A virus, working with @tompeacock.bsky.social and others between Liverpool, Pirbright, and Uppsala!

open to applications until next week!

Some areas of the text also seem formatted in a way less like a research paper and more like, say, a chatbot output..

It's Trans Awareness Week and @rssdiversity.bsky.social and @rainbowr.org's event tomorrow! 🏳️‍⚧️

Please join me from 17.30GMT at www.youtube.com/@RSSDiversity (stream will be up on homepage) to hear from Dr Voigt and Dr Webster on data on trans identities, barriers in trans lives, and better advocacy.

In two weeks! Live stand-up comedy from 100% real in-the-flesh researchers and scientists including me (am real last time I checked). This time we got everything from Scottish philosophy & saintspeak to astrophysics & mystery diseases!

www.thestand.co.uk/performance/...

👀 discount code: BRIGHTCLUB

This stack is able to correctly predict zoonotic potential of sequences in entirely unseen subtypes with AUC=0.95 and F1=0.90, a level of generalisability that is not often seen for machine learning host predictors.

Interestingly, it flags some duck H4 viruses from Americas as having distinct risk.

Training on 12 feature sets over each of 8 segments, we find protein properties are usually best at estimating zoonotic potential from a single segment.

But what about whole genomes? We can combine the best models in a single trained meta-learner (or "stack"), that draws on info from all of them!

We extracted ~19000 influenza sequences from birds and ~600 zoonotic sequences from humans (only non-seasonal subtypes).

Before training, we remove redundancy by grouping similar sequences into clusters. This is important to reduce bias, as most come from just a few subtypes like H7N9 and H5N1.

Lots of ML models can predict human spillover. However for influenza this task is harder because of a) genome segmentation, and b) strong signal within subtype or lineage.

We planned a model training architecture to handle this, ensuring predictions are rooted in virus biology, not shared ancestry.

If I had a nickel for every consecutive year we've had janky 3D-rendered big cats, I'd only have two nickels but it's weird it's happened twice.

I'm only able to join ViBioM virtually (thank you @evbc.bsky.social for making this available!!) but some fantastic talks from the @systemsvirology.bsky.social lab who are innovately retraining protein language models to infer virus evolution, antigenicity, and epi! おめでとうございます

Plane crashes, you say...

Scicomm update #2!

Bright Club Glasgow is looking for performers for our scientifically-hilarious comedy show in March!
Great way to develop new outreach skills and rethink how you talk research!

Get in touch: www.facebook.com/brightclubgl...

Scicomm update #1!

🛡️ Why did the military protect against dangers they never saw?
🔍 Can you trust evidence that seems to solve the murder trial?
🥪 And how risky is a bacon sandwich??

Find out in my free #stats-comm show in a week via Glasgow Humanists, Sun 9th Feb! www.meetup.com/glasgow-huma...

Kind of wild that we're basically at tier 5 of the Sid Meier's Alpha Centauri tech tree already

Rocky Horror and adult lunchables, Happy Christmas to me