Scholarlink

<p>This Data Descriptor presents a large-scale molecular descriptor dataset comprising 33,715 small molecules, specifically curated for high-throughput screening and drug delivery strategy optimization. Addressing the limitations of traditional descriptors in characterizing complex molecular geometry, we provide advanced descriptor families that encode spatial features, including bond and dihedral angles. These features are systematically captured through an operator-based matrix encoding strategy, utilizing a series of mathematical operations applied to atomic properties and graph-theoretical distances. The primary data record is provided as a 215.9 MB tab-separated text file, processed via a custom PHP-based pipeline to ensure standardized identifier mapping. This dataset is designed to enhance the predictive accuracy of inhibitor-target interaction models within computational drug discovery pipelines. By integrating high-dimensional structural data into an open-access format, this resource provides a robust framework for investigating the molecular mechanisms underlying drug efficacy.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A molecular descriptor dataset of 33,715 small molecules based on operator-based descriptor framework

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月21日

查看全文
  • Alexandra Farcas,
  • Lorentz Jäntschi

摘要

This Data Descriptor presents a large-scale molecular descriptor dataset comprising 33,715 small molecules, specifically curated for high-throughput screening and drug delivery strategy optimization. Addressing the limitations of traditional descriptors in characterizing complex molecular geometry, we provide advanced descriptor families that encode spatial features, including bond and dihedral angles. These features are systematically captured through an operator-based matrix encoding strategy, utilizing a series of mathematical operations applied to atomic properties and graph-theoretical distances. The primary data record is provided as a 215.9 MB tab-separated text file, processed via a custom PHP-based pipeline to ensure standardized identifier mapping. This dataset is designed to enhance the predictive accuracy of inhibitor-target interaction models within computational drug discovery pipelines. By integrating high-dimensional structural data into an open-access format, this resource provides a robust framework for investigating the molecular mechanisms underlying drug efficacy.