Summary of Learning Disentangled Equivariant Representation For Explicitly Controllable 3d Molecule Generation, by Haoran Liu et al.
Learning Disentangled Equivariant Representation for Explicitly Controllable 3D Molecule Generation
by Haoran Liu, Youzhi Luo, Tianxiao Li, James Caverlee, Martin Renqiang Min
First submitted to arxiv on: 19 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary We investigate the conditional generation of 3D drug-like molecules with explicit control over molecular properties such as Quantitative Estimate of Druglikeness or Synthetic Accessibility score, and effective binding to specific protein sites. Our proposed E(3)-equivariant Wasserstein autoencoder factorizes the latent space into disentangled aspects: molecular properties and structural context. This model ensures explicit control over attributes while maintaining equivariance and invariance. We introduce a novel alignment-based coordinate loss for auto-regressive de-novo 3D molecule generation from scratch. Our model demonstrates effectiveness in property-guided and context-guided molecule generation, both for de-novo design and structure-based drug discovery against protein targets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about creating new medicines with specific properties that make them more likely to work well on the body. It uses a special kind of computer program to generate these medicine designs from scratch. The program can control what properties the medicine has, such as how easily it will bind to a target in the body. This could be useful for finding new treatments for diseases. |
Keywords
» Artificial intelligence » Alignment » Autoencoder » Latent space
