4.5 Article

Saagar-A New, Extensible Set of Molecular Substructures for QSAR/QSPR and Read-Across Predictions

期刊

CHEMICAL RESEARCH IN TOXICOLOGY
卷 34, 期 2, 页码 634-640

出版社

AMER CHEMICAL SOC
DOI: 10.1021/acs.chemrestox.0c00464

关键词

-

资金

  1. [HHSN273201700001C]

向作者/读者索取更多资源

Molecular structure-based predictive models offer a cost-effective and efficient alternative to animal testing, with the need for interpretable descriptors to provide chemistry-backed predictive reasoning. Saagar, a novel chemistry-aware substructure, outperformed publicly available fingerprint sets in extracting compounds with higher scaffold similarity, showcasing its ability to efficiently characterize diverse chemical collections.
Molecular structure-based predictive models provide a proven alternative to costly and inefficient animal testing. However, due to a lack of interpretability of predictive models built with abstract molecular descriptors they have earned the notoriety of being black boxes. Interpretable models require interpretable descriptors to provide chemistry-backed predictive reasoning and facilitate intelligent molecular design. We developed a novel set of extensible chemistry-aware substructures, Saagar, to support interpretable predictive models and read-across protocols. Performance of Saagar in chemical characterization and search for structurally similar actives for read-across applications was compared with four publicly available fingerprint sets (MACCS (166), PubChem (881), ECFP4 (1024), ToxPrint (729)) in three benchmark sets (MUV, ULS, and Tox21) spanning similar to 145 000 compounds and 78 molecular targets at 1%, 2%, 5%, and 10% false discovery rates. In 18 of the 20 comparisons, interpretable Saagar features performed better than the publicly available, but less interpretable and fixed-bit length, fingerprints. Examples are provided to show the enhanced capability of Saagar in extracting compounds with higher scaffold similarity. Saagar features are interpretable and efficiently characterize diverse chemical collections, thus making them a better choice for building interpretable predictive in silico models and read-across protocols.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.5
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据