arXiv cs.AI· Utku Umur Acikalin, Yingheng Wang, Goncalo J. Gouveia, Tyler J. Schwertfeger, Frank C. Schroeder, Carla P. Gomes·· 12 小时前AI 评分29
GEMS：将分子结构解析建模为异构体图上的学习式搜索
GEMS: Molecular Structure Elucidation as Learned Search over Isomer Graphs
AI 导读
GEMS 通过保持分子式和原子价态的 2-switch 键编辑对候选结构进行学习式精修，在排除目标结构的 PubChem 初始化下，三轮编辑在 NPLIB1 和 MassSpecGym 上分别达到 29.64%/50.81% 和 30.22%/41.33% 的 Hit@1/Hit@10，超过已发表基线。
正文 · 原文
Abstract:Identifying small-molecule structures from molecular formulas and tandem mass spectra remains a bottleneck in untargeted metabolomics and natural-product discovery. Even when database retrieval or de novo prediction misses the exact structure, the resulting candidates can provide starting points for exploring nearby structures through a few bond edits. We introduce GEMS (Guided Editing of Molecular Structures), which refines candidates through learned 2-switch bond edits that preserve the molecular formula and each atom's total valence. GEMS uses MIST, a pretrained spectrum-to-fingerprint model, to predict fingerprint probabilities once per spectrum. A neural policy compares each candidate's fingerprint with these predictions, links present features to the atoms and bonds that generate them, and scores all valid edits in one forward pass. The policy is trained on shortest edit paths between structures, so it can be pretrained on PubChem isomers without spectra and then fine-tuned with spectrum-predicted fingerprints. At inference, GEMS searches from multiple initial candidates matching the supplied molecular formula and ranks candidates by fingerprint agreement and multi-start consensus. With the target structure excluded from PubChem initialization, three edit rounds achieve Hit@1/Hit@10 of 29.64%/50.81% on NPLIB1 and 30.22%/41.33% on MassSpecGym, exceeding published baselines under comparable settings. On NPLIB1, GEMS achieves higher identification accuracy in about 3.3 minutes end to end, compared with 26.2 hours for our FRIGID run. GEMS also improves identification from generated candidates, including those refined by FRIGID. Experiments with exact target fingerprints identify fingerprint prediction error as a bottleneck in target recovery and final ranking. Deeper search recovers additional targets without retraining, with larger gains from weaker initial candidates.
| Comments: | 36 pages, 9 figures, 17 tables |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.23617 [cs.AI] |
| (or arXiv:2610.23617v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.23617 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Utku Umur Acikalin [view email]
[v1]
Fri, 16 Oct 2026 15:28:34 UTC (706 KB)
来源：arXiv cs.AI · arxiv.org