A new rotation-translation invariant molecular encoding and its use in Computer-Aided Drug-Discovery

一种新的旋转翻译不变分子编码及其在计算机辅助药物发现中的应用

• 批准号: 17F17051
• 负责人: 山西 芳裕
• 金额: $ 1.47万
• 依托单位: Kyushu Institute of Technology (2018)Kyushu University (2017)
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for JSPS Fellows
• 财政年份: 2017
• 资助国家: 日本
• 起止时间: 2017-07-26 至 2019-03-31
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-17F17051/
• 关键词: LBVS; QSAR; applicability domain

I have been working on computational methods on chemoinformatics research. I have created the "Distance-Based Boolean Applicability Domain". In ligand-based modeling, one must create a model but also an applicability domain for that model. Existing methods are complex, hard to implement or difficult to interpret. They require a threshold or embed an empiric constant. I proposed a trivial to interpret domain. In High Throughput Screening data set modeling experiments, our domain improves the performance and early retrieval of models, while improving the scaffold diversity among top-ranked active molecules. I have published two articles in peer reviewed journals and presented posters at international conferences. I have released open source programs and libraries for chemoinformatics research. I have done some modeling for protein targets of interest to our team.

我一直致力于化学信息学研究中的计算方法。我已经创建了“基于距离的布尔适用性域”。在基于配体的建模中，必须创建一个模型，还要为该模型创建一个适用域。现有的方法很复杂，难以实施或难以解释。它们需要一个阈值或嵌入一个经验常数。我提出了一个琐碎的解释域。在高通量筛选数据集建模实验中，我们的领域提高了模型的性能和早期检索，同时提高了顶级活性分子之间的支架多样性。我在同行评议的期刊上发表了两篇文章，并在国际会议上展示了海报。我已经发布了用于化学信息学研究的开源程序和库。我为我们团队感兴趣的蛋白质目标做了一些建模。

项目成果

Consensus queries in ligand-based virtual screening experiments.

• DOI: 10.1186/s13321-017-0248-5
• 发表时间: 2017-11-28
• 期刊: Journal of cheminformatics
• 影响因子: 8.6
• 作者: Berenger F;Vu O;Meiler J
• 通讯作者: Meiler J