Publikation
From Fragments to Geometry: A Unified Graph Transformer for Molecular Representation from Conformer Ensembles
Ho Minh Duy Nguyen; Trung Nguyen; Ha Thi Hong Le; Mai Thanh Nhat Truong; TrungTin Nguyen; Nhat Ho; Khoa D Doan; Duy Duong-Tran; Li Shen; Daniel Sonntag; James Zou; Mathias Niepert; Hyojin Kim; Jonathan E Allen
In: ICML 2025 Generative AI and Biology (GenBio) Workshop. International Conference on Machine Learning (ICML-2025), 2025.
Zusammenfassung
Designing and understanding molecules for biological applications requires models that can integrate rich structural information from both 2D molecular graphs and diverse 3D conformer ensembles. We introduce a fragment-aware, structure-guided graph transformer that enables scalable and expressive molecular modeling by aggregating multiple 3D conformers while incorporating fragment-level inductive biases from the 2D topology. Our approach employs a trainable attention-based fusion mechanism within a graph transformer to dynamically combine 2D and 3D representations, moving beyond static solvers and rigid fusion heuristics. This architecture enables fine-grained reasoning over chemically diverse molecules, including organocatalysts and transition-metal complexes. While originally developed for molecular property prediction, the method’s structure-aware and fragment-level modeling is readily applicable to generative molecular design, enabling downstream applications in drug discovery, reaction modeling, and AI-driven biological research. The model scales to large datasets and achieves state-of-the-art results across molecular property benchmarks, demonstrating its potential as a foundational component for generative AI in molecular science.