ToxLens：一种可复现的图学习框架，用于感知数据泄露、不确定性校准的分子毒性预测
ToxLens: A Reproducible Graph-Learning Framework for Leakage-Aware, Uncertainty-Calibrated Molecular Toxicity Prediction
浏览论文内容
中文总结 AI 辅助
该研究提出可复现的ToxLens图学习框架，用于11项毒性终点的分子毒性预测，通过感知数据泄露的划分等技术优化，在数据泄露控制测试集上优于基线方法，可生成结构假设。
中文摘要 AI 辅助
分子毒性预测越来越多地用于在实验测试前对化合物进行优先级排序，但当结构相关分子出现在训练集和测试集中时，传统基准性能可能会夸大实际效用。我们提出了ToxLens，这是一种可复现的多任务图学习框架，用于11项毒性终点，涵盖Ames致突变性、急性口服毒性、hERG抑制，以及Tox21核受体和应激反应测定。该工作流程结合了保守化学整理、球排除过滤、感知数据泄露的UMAP-HDBSCAN划分、通过晚期连接结合的并行图和全局特征编码器、带共形式预测集的温度缩放蒙特卡洛丢弃、应用域分析，以及带遮挡控制的SHAP引导毒性基团发现。在数据泄露控制的测试划分上，五种子的软投票集成达到了马修斯相关系数0.44、受试者工作特征曲线下面积0.83、精确召回曲线下面积0.58的成绩。在相同划分和基于验证的阈值选择协议下，它在全部11项终点上优于四种基于ECFP4的浅层基线。控制消融实验显示全局通路很重要，而晚期连接优于测试的门控和逐特征线性调制融合变体。共形式预测集揭示了特定终点在集合效率上的显著差异，且判别性和校准度随与训练域的相似度提升而改善。在固定的已发布Tox21挑战和TDA划分上重新训练，产生了有竞争力但非全面最优的性能。SHAP引导的遮挡和共识子图挖掘产生了模型衍生的结构假设，其中44个包含至少一个符合预定义反事实标准的实例。
英文摘要
Molecular toxicity prediction is increasingly used to prioritise compounds before experimental testing, but conventional benchmark performance can overstate practical utility when structurally related molecules occur across training and test folds. We introduce ToxLens, a reproducible multi-task graph-learning framework for 11 toxicity endpoints spanning Ames mutagenicity, acute oral toxicity, hERG inhibition, and Tox21 nuclear-receptor and stress-response assays. The workflow combines conservative chemical curation, sphere-exclusion filtering, a leakage-aware UMAP-HDBSCAN split, parallel graph and global-feature encoders joined by late concatenation, temperature-scaled Monte Carlo dropout with conformal-style prediction sets, applicability-domain analysis, and SHAP-guided toxicophore discovery with occlusion controls. On the leakage-controlled test fold, a five-seed soft-voting ensemble achieved a Matthews correlation coefficient score of 0.44, an area under the receiver operating characteristic curve score of 0.83, and an area under the precision-recall curve score of 0.58. It exceeded four ECFP4-based shallow baselines on all 11 endpoints under the same split and validation-based threshold-selection protocol. Controlled ablations showed that the global pathway was important, whereas late concatenation outperformed the tested gated and feature-wise linear modulation fusion variants. Conformal-style prediction sets revealed substantial endpoint-specific variation in set efficiency, and discrimination and calibration improved with similarity to the training domain. Retraining on fixed published Tox21 Challenge and TDA folds produced competitive, but not uniformly state-of-the-art, performance. SHAP-guided occlusion and consensus subgraph mining yielded model-derived structural hypotheses, 44 of which contained at least one occurrence that passed the predefined counterfactual criteria.
发表机构
- The University of Queensland（昆士兰大学）
- Baker Heart and Diabetes Institute（贝克心脏与糖尿病研究所）
机构由 AI 辅助整理，请以论文原文为准。
