English

An End-to-End Framework for Molecular Conformation Generation via Bilevel Programming

Machine Learning 2025-01-07 v2 Biomolecules

Abstract

Predicting molecular conformations (or 3D structures) from molecular graphs is a fundamental problem in many applications. Most existing approaches are usually divided into two steps by first predicting the distances between atoms and then generating a 3D structure through optimizing a distance geometry problem. However, the distances predicted with such two-stage approaches may not be able to consistently preserve the geometry of local atomic neighborhoods, making the generated structures unsatisfying. In this paper, we propose an end-to-end solution for molecular conformation prediction called ConfVAE based on the conditional variational autoencoder framework. Specifically, the molecular graph is first encoded in a latent space, and then the 3D structures are generated by solving a principled bilevel optimization program. Extensive experiments on several benchmark data sets prove the effectiveness of our proposed approach over existing state-of-the-art approaches. Code is available at https://github.com/MinkaiXu/ConfVAE-ICML21

Keywords

Cite

@article{arxiv.2105.07246,
  title  = {An End-to-End Framework for Molecular Conformation Generation via Bilevel Programming},
  author = {Minkai Xu and Wujie Wang and Shitong Luo and Chence Shi and Yoshua Bengio and Rafael Gomez-Bombarelli and Jian Tang},
  journal= {arXiv preprint arXiv:2105.07246},
  year   = {2025}
}

Comments

Accepted by ICML 2021

R2 v1 2026-06-24T02:08:34.873Z