Prediction of drug-target interactions with 3D structure information of target binding sites

• Item type: SIG Technical Reports(1)
• 公開日: 2018-06-06
• タイトル: Prediction of drug-target interactions with 3D structure information of target binding sites
• タイトル / 言語: en
• タイトル: Prediction of drug-target interactions with 3D structure information of target binding sites
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: BIO一般セッション
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Presently with School of Computing, Tokyo Institute of Technology／Presently with Education Academy of Computational Life Sciences, Tokyo Institute of Technology
• 著者所属: Presently with School of Computing, Tokyo Institute of Technology／Presently with Education Academy of Computational Life Sciences, Tokyo Institute of Technology
• 著者所属(英): en
• 著者所属(英): Presently with School of Computing, Tokyo Institute of Technology / Presently with Education Academy of Computational Life Sciences, Tokyo Institute of Technology
• 著者所属(英): en
• 著者所属(英): Presently with School of Computing, Tokyo Institute of Technology / Presently with Education Academy of Computational Life Sciences, Tokyo Institute of Technology
• 著者名: Ruoming, He; Takashi, Ishida
• 著者名(英): Ruoming, He; Takashi, Ishida
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Predicting drug-target interactions is an important step for drug design. Previous method to compare target pairwise similarities by comparing amino acid sequences is effective but containing limitation when dealing with remote homology sequences. Using 3D structure information is better since protein structures often decide the functions and the interaction modes of drug-target pairs. However, difficulties on getting 3D structures of target proteins make it tough to extract and analyze the binding site structures from the target protein structures. Moreover, rather than the whole structure, the binding site structure of a target decides more on the drugs it interacts with according to the hypothesis that targets with similar binding sites are easier to interact with the same drug. Thus, our approach applied target binding site similarities to represent target pairwise similarities by using homology search to get the 3D structures as well as extracting and comparing the binding sites structures. Finally, our method improved prediction accuracy compared with previous methods.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Predicting drug-target interactions is an important step for drug design. Previous method to compare target pairwise similarities by comparing amino acid sequences is effective but containing limitation when dealing with remote homology sequences. Using 3D structure information is better since protein structures often decide the functions and the interaction modes of drug-target pairs. However, difficulties on getting 3D structures of target proteins make it tough to extract and analyze the binding site structures from the target protein structures. Moreover, rather than the whole structure, the binding site structure of a target decides more on the drugs it interacts with according to the hypothesis that targets with similar binding sites are easier to interact with the same drug. Thus, our approach applied target binding site similarities to represent target pairwise similarities by using homology search to get the 3D structures as well as extracting and comparing the binding sites structures. Finally, our method improved prediction accuracy compared with previous methods.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA12055912
• 書誌情報: 研究報告バイオ情報学（BIO）; 巻 2018-BIO-54, 号 44, p. 1-6, 発行日 2018-06-06
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8590
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会