HydroSophyTech/ToxiMol-benchmark 已完整同步
ToxiMol: A Benchmark for Structure-Level Molecular Detoxification
🔥 News
- 🏆 ToxiMol has been accepted as an Oral presentation at the 32nd SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2026), AI for Sciences Track.
Overview
ToxiMol is the first comprehensive benchmark for molecular toxicity repair tailored to general-purpose Multimodal Large Language Models (MLLMs). This is the dataset repository for the paper "Breaking Bad Molecules: Are MLLMs Ready for Structure-Level Molecular Detoxification?".
Key Features
🧬 Comprehensive Dataset
- 660 representative toxic molecules spanning diverse toxicity mechanisms and varying granularities
- 11 primary toxicity repair tasks based on Therapeutics Data Commons (TDC) platform
- Multi-granular coverage: Tox21 (12 sub-tasks), ToxCast (10 sub-tasks), and 9 additional datasets
- Multimodal inputs: SMILES strings + 2D molecular structure images rendered using RDKit
🎯 Challenging Task Definition
The molecular toxicity repair task requires models to:
- Identify potential toxicity endpoints from molecular structures
- Interpret semantic constraints from natural language descriptions
- Generate structurally similar substitute molecules that eliminate toxic fragments
- Satisfy drug-likeness and synthetic feasibility requirements
📊 Systematic Evaluation
- ToxiEval framework: Automated evaluation integrating toxicity prediction, synthetic accessibility, drug-likeness, and structural similarity
- Comprehensive analysis: Evaluation of ~30 mainstream MLLMs with ablation studies
- Multi-dimensional metrics: Success rate analysis across different evaluation criteria and failure modes
Dataset Structure
Task Overview
| Dataset | Task Type | # Molecules | Description |
|---|---|---|---|
| AMES | Binary Classification | 60 | Mutagenicity testing |
| Carcinogens | Binary Classification | 60 | Carcinogenicity prediction |
| ClinTox | Binary Classification | 60 | Clinical toxicity data |
| DILI | Binary Classification | 60 | Drug-induced liver injury |
| hERG | Binary Classification | 60 | hERG channel inhibition |
| hERG_Central | Binary Classification | 60 | Large-scale hERG database with integrated cardiac safety profiles |
| hERG_Karim | Binary Classification | 60 | hERG data |
15 个文件
浏览文件数据集版权信息
本数据集的许可证为 MIT License。如有违反相关条款,请联系 WEHUB,我们将及时处理。 查看许可证
下载数据集
通过 WeHub CLI 下载当前数据集快照。下列命令会固定为当前页面展示的数据版本(如果页面提供版本)。文件字节由本机直连存储下载,浏览器不会签发或保存下载链接。
前置要求
需要 Node.js 18 及以上,以及 npm(或 npx)。
1. 安装 CLI
npm install -g wehub-cli@latest
2. 下载此数据集
wehub datasets download ds_ext_2260_9adacf9832 --revision dedfdb2f898c0d8f30b35b03d604159179655c5f --output ./ds_ext_2260_9adacf9832
若中断或部分失败,在同一目录重新执行同一命令即可续传。默认会校验 SHA-256。
免全局安装
npx --yes wehub-cli@latest datasets download ds_ext_2260_9adacf9832 --revision dedfdb2f898c0d8f30b35b03d604159179655c5f --output ./ds_ext_2260_9adacf9832
高级选项
以下为 wehub datasets download 已支持的参数示例:
强制重新下载,不复用已校验的本地文件
wehub datasets download ds_ext_2260_9adacf9832 --revision dedfdb2f898c0d8f30b35b03d604159179655c5f --output ./ds_ext_2260_9adacf9832 --overwrite
仅包含匹配路径
wehub datasets download ds_ext_2260_9adacf9832 --revision dedfdb2f898c0d8f30b35b03d604159179655c5f --output ./ds_ext_2260_9adacf9832 --include "*.jsonl"
排除匹配路径
wehub datasets download ds_ext_2260_9adacf9832 --revision dedfdb2f898c0d8f30b35b03d604159179655c5f --output ./ds_ext_2260_9adacf9832 --exclude "*.md"
提高并发下载数
wehub datasets download ds_ext_2260_9adacf9832 --revision dedfdb2f898c0d8f30b35b03d604159179655c5f --output ./ds_ext_2260_9adacf9832 --jobs 8
完整帮助:wehub datasets download --help