Ke Yuan (@keyuan1) 's Twitter Profile
Ke Yuan

@keyuan1

Senior Lecturer @GlasgowCS @UofGlasgow | Machine Learning & Computational Biology

ID: 784487821

linkhttps://www.gla.ac.uk/schools/computing/staff/keyuan/ calendar_today27-08-2012 11:49:06

320 Tweet

377 Followers

934 Following

Craig Macdonald (@craig_macdonald) 's Twitter Profile Photo

Great to be involved in this work, applying information retrieval techniques to other types of language models - we show that cross-encoders can be used to predict if proteins will interact. We also apply our PLM-interact model for predicting virus-host interactions

Biology+AI Daily (@biologyaidaily) 's Twitter Profile Photo

PLM-interact: extending protein language models to predict protein-protein interactions 1. PLM-interact introduces a novel approach to predict protein-protein interactions (PPIs) by jointly encoding protein pairs, leveraging a method similar to “next sentence prediction” in NLP.

PLM-interact: extending protein language models to predict protein-protein interactions

1. PLM-interact introduces a novel approach to predict protein-protein interactions (PPIs) by jointly encoding protein pairs, leveraging a method similar to “next sentence prediction” in NLP.
DailyHealthcareAI (@aipulserx) 's Twitter Profile Photo

Can protein language models be adapted to accurately predict protein-protein interactions across different species and mutation scenarios? University of Glasgow bioRxiv "PLM-interact: extending protein language models to predict protein-protein interactions" • The prediction

Can protein language models be adapted to accurately predict protein-protein interactions across different species and mutation scenarios? <a href="/UofGlasgow/">University of Glasgow</a> <a href="/biorxivpreprint/">bioRxiv</a> 

"PLM-interact: extending protein language models to predict protein-protein interactions"

• The prediction
Ulad Litvin (@uladlitvin) 's Twitter Profile Photo

Our new viral protein structure database 🦠 Explore protein structure, function & evolution 🧬 Dive in 🤿: viro3d.cvr.gla.ac.uk

UofG MVLS (@uofgmvls) 's Twitter Profile Photo

🚀 MVLS #CollegeFutures PhD Programme! 🎓 Starting Oct 2025, we're offering 12 funded PhD studentships in: 🔬 Fundamentals of Life 💊 Advanced Diagnostics & Therapeutics 🌍 Planetary Health 💉 Health Inequalities Head of MVLS, Iain McInnes, explains our #MVLS2033 vision 📺

Geoff Macintyre (@gjmacintyre) 's Twitter Profile Photo

We are hiring! Opportunity for a project manager to oversee all aspects of a multi-centre clinical research project to implement an AI-based genomic test that predicts resistance to chemotherapy - Madrid, Spain. PM for more details. Deadline 25 January! cnio.es/en/empleo/resa…

CRUK Scotland Inst (@cruk_si) 's Twitter Profile Photo

📢Mark your diaries for the Beatson International Cancer Conference – "Cancer Models: from Data to Discovery" #BICC2025, returning on Tuesday 27th – Thursday 29th May 2025 at the Glasgow Science Centre! 📅Register and submit your abstract by 21 March: beatsonconference.org

📢Mark your diaries for the Beatson International Cancer Conference – "Cancer Models: from Data to Discovery" #BICC2025, returning on Tuesday 27th – Thursday 29th May 2025 at the Glasgow Science Centre!

📅Register and submit your abstract by 21 March: beatsonconference.org
CRUK Scotland Inst (@cruk_si) 's Twitter Profile Photo

🆕Opportunities to join a newly formed Data Science Team at the CRUK Scotland Institute focused on the management and sharing of multimodal data arising from pre-clinical and clinical datasets 📅Closing date: 28 February 👉Find out more and apply here: crukscotlandinstitute.ac.uk/careers

🆕Opportunities to join a newly formed Data Science Team at the CRUK Scotland Institute focused on the management and sharing of multimodal data arising from pre-clinical and clinical datasets

📅Closing date: 28 February 

👉Find out more and apply here: crukscotlandinstitute.ac.uk/careers
MRC-Uni of Glasgow Centre for Virus Research (@cvrinfo) 's Twitter Profile Photo

Congratulations Dan! 🎉 Dan Liu 's thesis focuses on leveraging pre-trained protein language models to predict virus-host associations and training large language models to predicted protein-protein interactions within and across species.

Congratulations Dan! 🎉

<a href="/DanLiu_/">Dan Liu</a> 's thesis focuses on leveraging pre-trained protein language models to predict virus-host associations and training large language models to predicted protein-protein interactions within and across species.
fajie yuan (@duguyuan) 's Twitter Profile Photo

Jin recently set up a Slack group for ColabSaprot discussions. Feel free to join here: westlakeai.slack.com/?redir=%2Farch… We have recently received positive experimental results from over 10 wet labs by using ColabSaprot. Video Tutorial: youtube.com/watch?v=nmLtjl…

GAMA Miguel Angel 🐦‍⬛🔑 (@miangoar) 's Twitter Profile Photo

Quick tip for busy people: you can improve the aesthetic of your proteins using the following commands in #ChimeraX. You can also add some shadows using the visual effects available in PowerPoint <protein emoji> ps. before the render, remove the struts using the visual display

Quick tip for busy people: you can improve the aesthetic of your proteins using the following commands in #ChimeraX. You can also add some shadows using the visual effects available in PowerPoint &lt;protein emoji&gt;

ps. before the render, remove the struts using the visual display
Sasha Rush (@srush_nlp) 's Twitter Profile Photo

Observation (for those not following): Universities in China have become more competitive at attracting junior academic researchers recently through facilitating corporate GPU use.

Amirhossein Kazemnejad (@a_kazemnejad) 's Twitter Profile Photo

Introducing nanoAhaMoment: Karpathy-style, single file RL for LLM library (<700 lines) - super hackable - no TRL / Verl, no abstraction💆‍♂️ - Single GPU, full param tuning, 3B LLM - Efficient (R1-zero countdown < 10h) comes with a from-scratch, fully spelled out YT video [1/n]

Introducing nanoAhaMoment: Karpathy-style, single file RL for LLM library (&lt;700 lines)

- super hackable
- no TRL / Verl, no abstraction💆‍♂️
- Single GPU, full param tuning, 3B LLM
- Efficient (R1-zero countdown &lt; 10h)

comes with a from-scratch, fully spelled out YT video [1/n]