Designing Cyclic Peptides via Harmonic SDE with Atom-Bond Modeling
International Conference on Machine Learning (ICML) 2025
Designing diverse cyclic peptides via full-atom diffusion modeling.
Research
Publications
2025 4
Orientation-Aware Graph Neural Networks for Protein Structure Representation Learning
International Conference on Research in Computational Molecular Biology (RECOMB) 2025
A new type of vector-based neural network combines directional 3D weights with equivariant message passing.
2024 5
Enhancing Protein Mutation Effect Prediction through a Retrieval-Augmented Framework
Conference and Workshop on Neural Information Processing Systems (NeurIPS) 2024
A retrieval-augmented model to improve protein mutation predictions by using local structure motif embeddings.
Categorical Flow Matching on Statistical Manifolds
Conference and Workshop on Neural Information Processing Systems (NeurIPS) 2024
A novel flow matching framework on the manifold of parameterized probability measures for discrete generation.
Full-Atom Peptide Design based on Multi-modal Flow Matching
International Conference on Machine Learning (ICML) 2024
The first multi-modal multi-task flow matching method for target-specific full-atom peptide design and analysis.
FAFE: Immune Complex Modeling with Geodesic Distance Loss on Noisy Group Frames
International Conference on Machine Learning (ICML) 2024 Spotlight
Replacing AF2's FAPE loss with SE(3) group-aware geodesic loss.
2022 2
Proximal Exploration for Model-guided Protein Sequence Design
International Conference on Machine Learning (ICML) 2022 Spotlight
Addressing model-guided sequence design using evolutionary search to find high-fitness mutants with low mutation counts.
Equivariant Point Cloud Analysis via Learning Orientations for Message Passing
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022 Oral
Learning virtual orientations for each point. Projecting neighbor points to local orientations before aggregating information from them.