🧪 Just out in Bioinformatics Advances: “Structure-based metabolite function prediction using graph neural networks”    Explore the full study here: https://doi.org/10.1093/bioadv/vbaf174 Authors include: @alisaadatv.bsky.social, @jannahastings.bsky.social

New Review Paper Alert! Are you interested in how to use AI to decode RNA? Check it out here www.nature.com/articles/s41... @jannahastings.bsky.social @vincentjung.bsky.social

Please share: @robertjwest.bsky.social @susanmichie.bsky.social @jannahastings.bsky.social @carolinacsilva.bsky.social @mayabraun.bsky.social @martamarques.bsky.social @mindandmatter.bsky.social @ucl-cbc.bsky.social

🚨 Registrations are now OPEN for the official launch of the APRICOT Project! 🚨 Join us as we explore how ontologies can advance behaviour change science. 🌍🧠 🗓️ 17 June 2025 ⏰ 3:30 PM – 5:00 PM (UK Time) Register here: tinyurl.com/HBCP-APRICOT

💡 Combine symbolic & neural AI to tackle biomedical challenges! Attend this tuto to use knowledge graphs + LLMs for evidence-based discovery with real cases in antibiotics & oncology 🔹Taught by Maxime Delmas & Andre Freitas (@idiap.bsky.social) 📅 Mon 8 Sep at #bc2 👉Register: tinyurl.com/5n8w3pwb

Hello from the APRICOT Full Team Meeting! 🌏 We’re deep in discussion and planning some exciting things. 👀 Stay tuned, exciting updates are coming your way soon!

C3NLP Workshop #NAACL2025: @iaugenstein.bsky.social will be presenting on tailoring LLM outputs to cultures, including where implicit cultural personalisation based on names leads to over-simplification arxiv.org/abs/2502.11995 @rnv.bsky.social @frimelle.bsky.social bsky.app/profile/frim... #NLProc

The SIB Swiss Institute of Bioinformatics (@sib.swiss) is proud to be organizing the #ECCB2026 in Geneva! 💪 🤔 New to SIB? We provide essential databases, software platforms & services to academia, companies, and hospitals. Learn more: www.sib.swiss 🔜Stay tuned to discover the preliminary programme

I mean… Yeah. Yeah they do. For entirely known and warned-about reasons. Yes. www.science.org/content/arti...

Come work with us! 😎 For the fantastic SNF-ANR collaborative @metabolinkai.bsky.social project, we are looking for a computational PhD student based in Zurich. If you are excited by knowledge-based approaches to AI and metabolomics data, we want to hear from you. To apply jobs.uzh.ch/job-vacancie...

Our third and final #AI webinar is also open for registration. In this webinar our speakers @jannahastings.bsky.social, @susanmichie.bsky.social & @robertjwest.bsky.social will explore ethical challenges in AI for behavioural research. 📅 1pm, 12 May 2025. us02web.zoom.us/webinar/regi...

🎆 🎉 Great news: Amos Bairoch has been selected to receive the 2025 ISCB Accomplishment by a Senior Scientist Award! Join us in congratulating him 👏 !

[BC]² Basel Computational Biology Conference 2025 will take place on 8-10 September. The conference brings together over 500 international scientists from academia, industry, and healthcare to explore advancements at the intersection of #bioinformatics and artificial intelligence. See bc2.ch

Check out our recently published paper about evaluating AI-generated images! Wonderful work together with @jannahastings.bsky.social @muhrlizza.bsky.social and more! www.cureus.com/articles/308...

Join us for BR-UK’s first webinar of 2025 where @jannahastings.bsky.social, @susanmichie.bsky.social and @robertjwest.bsky.social will discuss Artificial Intelligence in Behavioural Research. 📅 28 January 2025 🕦 1:30pm Book to watch live via Zoom: us02web.zoom.us/webinar/regi...

🚨 Upcoming webinar - Join our Director @susanmichie.bsky.social, and CBC Associates @robertjwest.bsky.social & @jannahastings.bsky.social, to explore the ethical challenges in applying AI to behavioural research & potential solutions. More details & link to register below 👇

BR-UK's January 2025 webinar on artificial intelligence in behavioural research is open for registration now.📅 🎙With speakers @jannahastings.bsky.social, @susanmichie.bsky.social and @robertjwest.bsky.social us02web.zoom.us/webinar/regi...

Three Lecturer openings at @citystgeorges.bsky.social: www.city.ac.uk/about/jobs/a.... Join our new AI center covering, among others: NeSy AI, MAS, Reinforcement Learning, NLP, Computer Vision, KGs, Adaptive Systems. @arturgarcez.bsky.social @pmadhyastha.bsky.social @ealonso.bsky.social

I'm sure this Science story on how systematic reviews are threatened by paper mills is good, but the illustration by Sara Gironi Carnevale is great saragironicarnevale.com Story: https://www.science.org/content/article/systematic-reviews-aim-extract-broad-conclusions-many-studies-are-peril

Just finished reading Stanislaw Lem's "Eden" which was very very good. The first of his many "social sci fi" novels about failed first contact and the difficulty of communication. Reminded me of Annihilation and Scavengers Reign. Also I love all the vintage covers and need to collect them all.

Need a prompt version of: imgs.xkcd.com/comics/explo...

Today, the University of Basel celebrated its Dies academicus and awarded honorary doctorates to seven [WHITE MALE] individuals for their services to science and society. 😢 #EDI

I don't want to cast doubt on the impeccable dependability of google AI, but I don't remember this particular plot twist in du Maurier's REBECCA

Imagine a snail. Imagine 5 snails. Imagine a snail on a sail. Imagine something that isn’t a snail. now read a new pre-print: "Relations, Negations, & Numbers: Looking for Logic in Generative Text-to-Image Models" (by Conwell, Tawiah-Quashie, & me) arxiv.org/pdf/2411.17066

A growing problem --- fake papers polluting systematic reviews. And interesting point that the gold-standard protocols that Cochrane is developing to weed out untrustworthy studies, are too much of an effort for systematic-review authors looking at 100s of papers. www.science.org/content/arti...

🩺💡The Bitterman lab has spent much of the past year researching #LLMs for healthcare. This post summarizes our inroads into making LLMs safer and reliable for clinicians and patients: huggingface.co/blog/shanche.... We'll be at #EMNLP2024 - come chat if you have similar interests!

3.5/🧵 Moreover, alignment methods don’t resolve inconsistencies in disease prevalence across languages (EN 🇺🇸, ES 🇪🇸, FR 🇫🇷, ZH 🇨🇳). And tuning on English usually only affects English prompt output

Crosscare is accepted @neuripsconf.bsky.social 🎉 We showed LLMs are far from grounded with true prevalence, and groundings across languages are so inconsistent! Also, a dashboard for people to explore the prevalence data across diseases and racial groups: crosscare.net #NeurIPS2024

The main question about the current LLM “reasoning” research is what to do next. Most go into synthetic generation and training on maybe with self-Refinement in hopes the model becomes better. I think we are missing controlled task formalization, step by step reasoning and strict step verification.

I was so proud to receive this year’s Society for the Study of Addiction Impact Prize for my work on reducing inequalities through smoking cessation interventions. The work I do isn’t particularly ‘scientifically sexy’ but it’s important and I am so pleased that’s recognised.