Proceedings

Equivariant Graph Neural Networks for Protein Interaction Modeling and Structure-Aware Molecular Design

Animesh Animesh

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Doctoral Consortium. Pages 8319-8320.

https://doi.org/10.24963/ijcai.2026/938

Keywords

  • Machine Learning: Attention models
  • Multidisciplinary Topics and Applications: Life sciences
  • Machine Learning: Geometric learning

Abstract

Equivariant graph neural networks (EGNNs) have revolutionized molecular property prediction by encoding 3D structures while preserving physical symmetry. However, existing EGNNs struggle to capture fine-grained geometric features needed for tasks like understanding protein interactions, and designing molecules that modulate them are central challenges in computational drug discovery.

This thesis develops a progressive framework based on equivariant graph neural networks (EqGNNs) that respects the geometric symmetries of molecular structures.

Our contributions progress along two axes: from discriminative to generative and from basic representations to physics-informed, language-conditioned generation. In this study, we propose novel methods for both discriminative (E(Q)AGNN-PPIS, GDEGAN) and generative tasks (PhysFlow). Together, these advance symmetry-aware geometric deep learning for drug discovery.

Animesh Animesh. 2026. Equivariant Graph Neural Networks for Protein Interaction Modeling and Structure-Aware Molecular Design. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 8319–8320. https://doi.org/10.24963/ijcai.2026/938
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards