TrimNet : learning molecular representation from triplet messages for biomedicine
© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissionsoup.com..
MOTIVATION: Computational methods accelerate drug discovery and play an important role in biomedicine, such as molecular property prediction and compound-protein interaction (CPI) identification. A key challenge is to learn useful molecular representation. In the early years, molecular properties are mainly calculated by quantum mechanics or predicted by traditional machine learning methods, which requires expert knowledge and is often labor-intensive. Nowadays, graph neural networks have received significant attention because of the powerful ability to learn representation from graph data. Nevertheless, current graph-based methods have some limitations that need to be addressed, such as large-scale parameters and insufficient bond information extraction.
RESULTS: In this study, we proposed a graph-based approach and employed a novel triplet message mechanism to learn molecular representation efficiently, named triplet message networks (TrimNet). We show that TrimNet can accurately complete multiple molecular representation learning tasks with significant parameter reduction, including the quantum properties, bioactivity, physiology and CPI prediction. In the experiments, TrimNet outperforms the previous state-of-the-art method by a significant margin on various datasets. Besides the few parameters and high prediction accuracy, TrimNet could focus on the atoms essential to the target properties, providing a clear interpretation of the prediction tasks. These advantages have established TrimNet as a powerful and useful computational tool in solving the challenging problem of molecular representation learning.
AVAILABILITY: The quantum and drug datasets are available on the website of MoleculeNet: http://moleculenet.ai. The source code is available in GitHub: https://github.com/yvquanli/trimnet.
CONTACT: xjyaolzu.edu.cn, songsen@tsinghua.edu.cn.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2021 |
|---|---|
| Erschienen: |
2021 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:22 |
|---|---|
| Enthalten in: |
Briefings in bioinformatics - 22(2021), 4 vom: 20. Juli |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Li, Pengyong [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Compound–protein interaction |
|---|
| Anmerkungen: |
Date Completed 19.11.2021 Date Revised 19.11.2021 published: Print Citation Status MEDLINE |
|---|
| doi: |
10.1093/bib/bbaa266 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM317158333 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM317158333 | ||
| 003 | DE-627 | ||
| 005 | 20231225162642.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2021 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1093/bib/bbaa266 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1057.xml |
| 035 | |a (DE-627)NLM317158333 | ||
| 035 | |a (NLM)33147620 | ||
| 035 | |a (PII)bbaa266 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Li, Pengyong |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a TrimNet |b learning molecular representation from triplet messages for biomedicine |
| 264 | 1 | |c 2021 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 19.11.2021 | ||
| 500 | |a Date Revised 19.11.2021 | ||
| 500 | |a published: Print | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © The Author(s) 2020. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissionsoup.com. | ||
| 520 | |a MOTIVATION: Computational methods accelerate drug discovery and play an important role in biomedicine, such as molecular property prediction and compound-protein interaction (CPI) identification. A key challenge is to learn useful molecular representation. In the early years, molecular properties are mainly calculated by quantum mechanics or predicted by traditional machine learning methods, which requires expert knowledge and is often labor-intensive. Nowadays, graph neural networks have received significant attention because of the powerful ability to learn representation from graph data. Nevertheless, current graph-based methods have some limitations that need to be addressed, such as large-scale parameters and insufficient bond information extraction | ||
| 520 | |a RESULTS: In this study, we proposed a graph-based approach and employed a novel triplet message mechanism to learn molecular representation efficiently, named triplet message networks (TrimNet). We show that TrimNet can accurately complete multiple molecular representation learning tasks with significant parameter reduction, including the quantum properties, bioactivity, physiology and CPI prediction. In the experiments, TrimNet outperforms the previous state-of-the-art method by a significant margin on various datasets. Besides the few parameters and high prediction accuracy, TrimNet could focus on the atoms essential to the target properties, providing a clear interpretation of the prediction tasks. These advantages have established TrimNet as a powerful and useful computational tool in solving the challenging problem of molecular representation learning | ||
| 520 | |a AVAILABILITY: The quantum and drug datasets are available on the website of MoleculeNet: http://moleculenet.ai. The source code is available in GitHub: https://github.com/yvquanli/trimnet | ||
| 520 | |a CONTACT: xjyaolzu.edu.cn, songsen@tsinghua.edu.cn | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, Non-U.S. Gov't | |
| 650 | 4 | |a compound–protein interaction | |
| 650 | 4 | |a computational method | |
| 650 | 4 | |a deep learning | |
| 650 | 4 | |a graph neural networks | |
| 650 | 4 | |a molecular property | |
| 650 | 4 | |a molecular representation | |
| 700 | 1 | |a Li, Yuquan |e verfasserin |4 aut | |
| 700 | 1 | |a Hsieh, Chang-Yu |e verfasserin |4 aut | |
| 700 | 1 | |a Zhang, Shengyu |e verfasserin |4 aut | |
| 700 | 1 | |a Liu, Xianggen |e verfasserin |4 aut | |
| 700 | 1 | |a Liu, Huanxiang |e verfasserin |4 aut | |
| 700 | 1 | |a Song, Sen |e verfasserin |4 aut | |
| 700 | 1 | |a Yao, Xiaojun |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Briefings in bioinformatics |d 2000 |g 22(2021), 4 vom: 20. Juli |w (DE-627)NLM11366883X |x 1477-4054 |7 nnns |
| 773 | 1 | 8 | |g volume:22 |g year:2021 |g number:4 |g day:20 |g month:07 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1093/bib/bbaa266 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 22 |j 2021 |e 4 |b 20 |c 07 | ||