DMol: A Highly Efficient and Chemical Motif-Preserving Molecule Generation Platform
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915590568935424 |
|---|---|
| author | Niu, Peizhi Wang, Yu-Hsiang Rana, Vishal Rupakheti, Chetan Pandey, Abhishek Milenkovic, Olgica |
| author_facet | Niu, Peizhi Wang, Yu-Hsiang Rana, Vishal Rupakheti, Chetan Pandey, Abhishek Milenkovic, Olgica |
| contents | We introduce a new graph diffusion model for small molecule generation, DMol, which outperforms the state-of-the-art DiGress model in terms of validity by roughly 1.5% across all benchmarking datasets while reducing the number of diffusion steps by at least 10-fold, and the running time to roughly one half. The performance improvements are a result of a careful change in the objective function and a graph noise scheduling approach which, at each diffusion step, allows one to only change a subset of nodes of varying size in the molecule graph. Another relevant property of the method is that it can be easily combined with junction-tree-like graph representations that arise by compressing a collection of relevant ring structures into supernodes. Unlike classical junction-tree techniques that involve VAEs and require complicated reconstruction steps, compressed DMol directly performs graph diffusion on a graph that compresses only a carefully selected set of frequent carbon rings into supernodes, which results in straightforward sample generation. This compressed DMol method offers additional validity improvements over generic DMol of roughly 2%, increases the novelty of the method, and further improves the running time due to reductions in the graph size. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_06312 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DMol: A Highly Efficient and Chemical Motif-Preserving Molecule Generation Platform Niu, Peizhi Wang, Yu-Hsiang Rana, Vishal Rupakheti, Chetan Pandey, Abhishek Milenkovic, Olgica Machine Learning Artificial Intelligence We introduce a new graph diffusion model for small molecule generation, DMol, which outperforms the state-of-the-art DiGress model in terms of validity by roughly 1.5% across all benchmarking datasets while reducing the number of diffusion steps by at least 10-fold, and the running time to roughly one half. The performance improvements are a result of a careful change in the objective function and a graph noise scheduling approach which, at each diffusion step, allows one to only change a subset of nodes of varying size in the molecule graph. Another relevant property of the method is that it can be easily combined with junction-tree-like graph representations that arise by compressing a collection of relevant ring structures into supernodes. Unlike classical junction-tree techniques that involve VAEs and require complicated reconstruction steps, compressed DMol directly performs graph diffusion on a graph that compresses only a carefully selected set of frequent carbon rings into supernodes, which results in straightforward sample generation. This compressed DMol method offers additional validity improvements over generic DMol of roughly 2%, increases the novelty of the method, and further improves the running time due to reductions in the graph size. |
| title | DMol: A Highly Efficient and Chemical Motif-Preserving Molecule Generation Platform |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2504.06312 |