Andrew Foong

We made it on the cover! So proud of the team 🎉 You can use BioEmu for free now to get AI predictions for the entire distribution of conformations a protein can take: github.com/microsoft/bi...

Scalable emulation of protein equilibrium ensembles with generative deep learning Following the sequence and structure revolutions, predicting functionally relevant protein structure changes at scale remains an outstanding challenge. We introduce BioEmu, a deep learning system that...

Scalable emulation of protein equilibrium ensembles with generative deep learning

www.science.org/doi/10.1126/...

BioEmu is out! Grateful to have had the opportunity to work with such an incredible team on this project 🤗

Scalable emulation of protein equilibrium ensembles with generative deep learning Following the sequence and structure revolutions, predicting functionally relevant protein structure changes at scale remains an outstanding challenge. We introduce BioEmu, a deep learning system that...

BioEmu is now on @science.org ! The revised version includes an upgraded model and makes a lot of MD simulation data internally generated at MSR available to the public. This took a lot of firepower from us in the last two years.
www.science.org/doi/10.1126/...

Great to see this picked up by @natbiotech.nature.com!

An incredible multidisciplinary team effort with my ex-colleagues at Microsoft Research AI for Science. Proud to be part of it!

With this publication, we hope we further cement the idea that deep generative models can go beyond static structure generation and towards capturing biologically relevant dynamics at scale.

BioEmu is a generative deep learning model for emulating protein equilibrium ensembles. It's already been open-sourced and used across the community to generate high-fidelity protein structural ensembles.

Scalable emulation of protein equilibrium ensembles with generative deep learning Following the sequence and structure revolutions, predicting functionally relevant protein structure changes at scale remains an outstanding challenge. We introduce BioEmu, a deep learning system that...

🎉 Thrilled to share that our work on BioEmu has been published in Science: www.science.org/doi/10.1126/...

Today in the journal Science: BioEmu from Microsoft Research AI for Science. This generative deep learning method emulates protein equilibrium ensembles – key for understanding protein function at scale. www.science.org/doi/10.1126/...

BioEmu now published in @science.org !!

What is BioEmu? Check out this video:
youtu.be/LStKhWcL0VE?...

5/ At @mayoclinic.org, clinicians are at the forefront of AI innovation and deployment. Progress depends on a deep, practical understanding of how AI works. I hope this series helps bridge the gap between cutting-edge technology and clinical care. Let me know what you think and feel free to share!

4/ 💬 Lectures 5 & 6: How ChatGPT works, and how to use it.
- How ChatGPT was created
- Prompting strategies
- How to reduce hallucinations with RAG
If you're using LLMs in medicine, start here.

3/ 🖼️ Lectures 3 & 4: How AI sees medical images.
- What AI imaging models actually “look at”
- Convolutional neural networks (CNNs)
For anyone thinking of applying AI to image data.

2/ 🧠 Lectures 1 & 2: Foundational concepts of deep learning from the ground up.
- Gradient descent
- Neural networks
- Generalisation and overfitting
If you’re new to AI and want to grasp the basics, start here.

🧵1/ How does AI actually work under the hood? I just delivered a 6-part lecture series to clinician colleagues at @mayoclinic.org:
"Understanding AI from Scratch: From Linear Regression to ChatGPT"
Watch here: mssvideoupload.mayo.edu/playlist/ded...

I’ll be at ICLR 2025 in sunny Singapore! Contact me to chat about AI in healthcare and open positions at Mayo Clinic!

Life-Changing Careers at Mayo Clinic Learn more about applying for Research Fellow - AI & Data Analytics (AIDA), Radiation Oncology at Mayo Clinic

If this sounds like a fit — or if you're curious to learn more — here's the link: jobs.mayoclinic.org/job/rocheste...

We're looking for curious, technically strong researchers who want to work on problems that matter. Background in ML, stats, or applied data science is ideal. Radiation oncology experience welcome, but not required.

Having worked in both academic and industry labs, I’ve seen what each does best, and what they sometimes can’t.

With AIDA, we’re closing the loop between idea → data → model → clinical feedback → deployment — all under one roof.

Our aim is to bridge the gap between the ambition of leading AI labs, the rigor of academic research, and the grounded reality of patient care.

AIDA is a research group embedded directly within one of the world’s busiest cancer centers.

We work at the intersection of:
– Clinical data at scale
– Cutting-edge ML
– Daily oncology practice

We're hiring! The AI and Data Analytics (AIDA) team in Mayo Clinic Radiation Oncology is recruiting Research Fellows to advance machine learning for real-world clinical impact. 🧵