Cargando…
Improving Chemical Reaction Prediction with Unlabeled Data
Predicting products of organic chemical reactions is useful in chemical sciences, especially when one or more reactants are new organics. However, the performance of traditional learning models heavily relies on high-quality labeled data. In this work, to utilize unlabeled data for better prediction...
Autores principales: | , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9506495/ https://www.ncbi.nlm.nih.gov/pubmed/36144703 http://dx.doi.org/10.3390/molecules27185967 |
_version_ | 1784796737170309120 |
---|---|
author | Xie, Yu Zhang, Yuyang Wong, Ka-Chun Shi, Meixia Peng, Chengbin |
author_facet | Xie, Yu Zhang, Yuyang Wong, Ka-Chun Shi, Meixia Peng, Chengbin |
author_sort | Xie, Yu |
collection | PubMed |
description | Predicting products of organic chemical reactions is useful in chemical sciences, especially when one or more reactants are new organics. However, the performance of traditional learning models heavily relies on high-quality labeled data. In this work, to utilize unlabeled data for better prediction performance, we propose a method that combines semi-supervised learning with graph convolutional neural networks for chemical reaction prediction. First, we propose a Mean Teacher Weisfeiler–Lehman Network to find the reaction centers. Then, we construct the candidate product set. Finally, we use an Improved Weisfeiler–Lehman Difference Network to rank candidate products. Experimental results demonstrate that, with 400k labeled data, our framework can improve the top-5 accuracy by 0.7% using 35k unlabeled data. When the proportion of unlabeled data increases, the performance gain can be larger. For example, with 80k labeled data and 35k unlabeled data, the performance gain with our framework can be 1.8%. |
format | Online Article Text |
id | pubmed-9506495 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-95064952022-09-24 Improving Chemical Reaction Prediction with Unlabeled Data Xie, Yu Zhang, Yuyang Wong, Ka-Chun Shi, Meixia Peng, Chengbin Molecules Article Predicting products of organic chemical reactions is useful in chemical sciences, especially when one or more reactants are new organics. However, the performance of traditional learning models heavily relies on high-quality labeled data. In this work, to utilize unlabeled data for better prediction performance, we propose a method that combines semi-supervised learning with graph convolutional neural networks for chemical reaction prediction. First, we propose a Mean Teacher Weisfeiler–Lehman Network to find the reaction centers. Then, we construct the candidate product set. Finally, we use an Improved Weisfeiler–Lehman Difference Network to rank candidate products. Experimental results demonstrate that, with 400k labeled data, our framework can improve the top-5 accuracy by 0.7% using 35k unlabeled data. When the proportion of unlabeled data increases, the performance gain can be larger. For example, with 80k labeled data and 35k unlabeled data, the performance gain with our framework can be 1.8%. MDPI 2022-09-14 /pmc/articles/PMC9506495/ /pubmed/36144703 http://dx.doi.org/10.3390/molecules27185967 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Xie, Yu Zhang, Yuyang Wong, Ka-Chun Shi, Meixia Peng, Chengbin Improving Chemical Reaction Prediction with Unlabeled Data |
title | Improving Chemical Reaction Prediction with Unlabeled Data |
title_full | Improving Chemical Reaction Prediction with Unlabeled Data |
title_fullStr | Improving Chemical Reaction Prediction with Unlabeled Data |
title_full_unstemmed | Improving Chemical Reaction Prediction with Unlabeled Data |
title_short | Improving Chemical Reaction Prediction with Unlabeled Data |
title_sort | improving chemical reaction prediction with unlabeled data |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9506495/ https://www.ncbi.nlm.nih.gov/pubmed/36144703 http://dx.doi.org/10.3390/molecules27185967 |
work_keys_str_mv | AT xieyu improvingchemicalreactionpredictionwithunlabeleddata AT zhangyuyang improvingchemicalreactionpredictionwithunlabeleddata AT wongkachun improvingchemicalreactionpredictionwithunlabeleddata AT shimeixia improvingchemicalreactionpredictionwithunlabeleddata AT pengchengbin improvingchemicalreactionpredictionwithunlabeleddata |