Enhancing Ligand Property and Activity Prediction and Interpretation Using Multiple Molecular Graph Representations

Material type: 文書・図像類

Author: Kengkanna, Apakornほか

Publisher: 情報処理学会

Publication date: 2023-03

Material Format: Paper

Capacity, size, etc.: -

NDC: -

View All

Notes on use

Note (General)：: Graph neural networks (GNNs) are effective at predicting compound properties and activities, but they have two limitations: molecular representation a...

Search by Bookstore

Holdings of Libraries in Japan

This page shows libraries in Japan other than the National Diet Library that hold the material.

List of Cooperating Institutions and Databases

Please contact your local library for information on how to use materials or whether it is possible to request materials from the holding libraries.

other

Tokyo Tech Research Repository
Paper
You can check the holdings of institutions and databases with which 学術機関リポジトリデータベース（IRDB）（機関リポジトリ） is linked at the site of 学術機関リポジトリデータベース（IRDB）（機関リポジトリ）.
Tokyo Tech Research Repository

Search by Bookstore

Bibliographic Record

You can check the details of this material, its authority (keywords that refer to materials on the same subject, author's name, etc.), etc.

Paper

Material Type: 文書・図像類
Title: Enhancing Ligand Property and Activity Prediction and Interpretation Using Multiple Molecular Graph Representations
Author/Editor: Kengkanna, Apakorn
大上, 雅史
Ohue, Masahito
Author Heading: Kengkanna, Apakorn
大上, 雅史
Ohue, Masahito
Publication, Distribution, etc.: 情報処理学会
Publication Date: 2023-03
Publication Date (W3CDTF): 2023-03
Alternative Title: Enhancing Ligand Property and Activity Prediction and Interpretation Using Multiple Molecular Graph Representations
Periodical title: 研究報告バイオ情報学（BIO）
No. or year of volume/issue: 2023-BIO-73 35
Volume: 2023-BIO-73
Issue: 35
Pages: 1-4
ISSN (Periodical Title): ISSN : 2188-8590
Text Language Code: jpn
Target Audience: 一般
Note (General): Graph neural networks (GNNs) are effective at predicting compound properties and activities, but they have two limitations: molecular representation and interpretability. Atom-based molecular graphs are commonly used to represent molecules, but they may not capture important substructures/functional groups that highly impact molecular properties. Interpretability is also important as it can provide scientific insight for optimization, but GNNs are complex and less interpretable. This research introduces techniques to create alternative reduced molecular graph representations that integrate higher-level information and support interpretation. Experiments are conducted on pharmaceutical endpoint datasets, and attention mechanism is used to identify significant substructures. Results show that combining multiple graph representations gives promising performance and provides interpretation aligning with background knowledge. This research could facilitate model understanding and applications in drug discovery.
identifier:oai:t2r2.star.titech.ac.jp:50659460
Data Provider (Database): 国立情報学研究所 : 学術機関リポジトリデータベース（IRDB）（機関リポジトリ）
https://irdb.nii.ac.jp
Original Data Provider (Database): 東京工業大学 : 東京工業大学リサーチリポジトリ