First broadly generalizable diffusion model for 3D molecular generation. State-of-the-art results for periodic crystals and non-periodic molecular systems through transfer learning. (ICML 2025 & Spotlight at ICLR 2025 Workshop)

A new generative toolkit for inverse design of 3D RNA structure. gRNAde has been validated in wet labs, achieving expert-level accuracy and outperforming all previous computation methods at designing RNA structure and function. (ICLR 2025 Spotlight & invited book chapter for Methods in Molecular Biology)

A ‘Hitchhiker’s Guide’ introducing deep learning for molecular modelling—written without the goal of publication—if we could travel back in time and learn from it at the start of our PhDs! 🤗

A unified theory of Graph Neural Networks and Transformers for 3D molecular structure. And mechanistic understanding of how architectural decisions, such as invariance vs. equivariance, impact the class of functions that models can express. (ICML 2023 & Oral at NeurIPS 2022 Workshop)

A currated reading + watch list for beginners to get started with RNA modelling and desing.

Exploring the connection between Transformer models such as GPT and BERT for Natural Language Processing, and Graph Neural Networks. (80,000+ readers on The Gradient, featured in Probabilistic ML textbook and classes at Stanford, Cambridge, Oxford.)