A Deep Neural Network for Accurate and Robust Prediction of the Glass Transition Temperature of Polyhydroxyalkanoate Homo- and Copolymers
Abstract
The purpose of this study was to develop a data-driven machine learning model to predict the performance properties of polyhydroxyalkanoates (PHAs), a group of biosourced polyesters featuring excellent performance, to guide future design and synthesis experiments. A deep neural network (DNN) machine learning model was built for predicting the glass transition temperature, Tg, of PHA homo- and copolymers. Molecular fingerprints were used to capture the structural and atomic information of PHA monomers. The other input variables included the molecular weight, the polydispersity index, and the percentage of each monomer in the homo- and copolymers. The results indicate that the DNN model achieves high accuracy in estimation of the glass transition temperature of PHAs. In addition, the symmetry of the DNN model is ensured by incorporating symmetry data in the training process. The DNN model achieved better performance than the support vector machine (SVD), a nonlinear ML model and least absolute shrinkage and selection operator (LASSO), a sparse linear regression model. The relative importance of factors affecting the DNN model prediction were analyzed. Sensitivity of the DNN model, including strategies to deal with missing data, were also investigated. Compared with commonly used machine learning models incorporating quantitative structure–property (QSPR) relationships, itmore »
- Authors:
-
- Case Western Reserve Univ., Cleveland, OH (United States). Dept. of Civil and Environmental Engineering
- Case Western Reserve Univ., Cleveland, OH (United States). Depts. of Computer and Data Sciences and Electrical, Computer and Systems Engineering
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States). Bioscience Division
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States). Materials Science and Technology Division
- (Bill) [Case Western Reserve Univ., Cleveland, OH (United States). Depts. of Computer and Data Sciences and Electrical, Computer and Systems Engineering, and Dept. of Civil and Environmental Engineering
- Publication Date:
- Research Org.:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC)
- OSTI Identifier:
- 1815763
- Alternate Identifier(s):
- OSTI ID: 1861292
- Report Number(s):
- LA-UR-20-25907
Journal ID: ISSN 1996-1944; MATEG9; PII: ma13245701; TRN: US2213576
- Grant/Contract Number:
- 89233218CNA000001
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Materials
- Additional Journal Information:
- Journal Volume: 13; Journal Issue: 24; Journal ID: ISSN 1996-1944
- Publisher:
- MDPI
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; 36 MATERIALS SCIENCE; molecular fingerprint; deep neural network; glass transition temperature; copolymers; quantitative structure–property relationship (QSPR)
Citation Formats
Jiang, Zhuoying, Hu, Jiajie, Marrone, Babetta L., Pilania, Ghanshyam, and Yu, Xiong. A Deep Neural Network for Accurate and Robust Prediction of the Glass Transition Temperature of Polyhydroxyalkanoate Homo- and Copolymers. United States: N. p., 2020.
Web. doi:10.3390/ma13245701.
Jiang, Zhuoying, Hu, Jiajie, Marrone, Babetta L., Pilania, Ghanshyam, & Yu, Xiong. A Deep Neural Network for Accurate and Robust Prediction of the Glass Transition Temperature of Polyhydroxyalkanoate Homo- and Copolymers. United States. https://doi.org/10.3390/ma13245701
Jiang, Zhuoying, Hu, Jiajie, Marrone, Babetta L., Pilania, Ghanshyam, and Yu, Xiong. Mon .
"A Deep Neural Network for Accurate and Robust Prediction of the Glass Transition Temperature of Polyhydroxyalkanoate Homo- and Copolymers". United States. https://doi.org/10.3390/ma13245701. https://www.osti.gov/servlets/purl/1815763.
@article{osti_1815763,
title = {A Deep Neural Network for Accurate and Robust Prediction of the Glass Transition Temperature of Polyhydroxyalkanoate Homo- and Copolymers},
author = {Jiang, Zhuoying and Hu, Jiajie and Marrone, Babetta L. and Pilania, Ghanshyam and Yu, Xiong},
abstractNote = {The purpose of this study was to develop a data-driven machine learning model to predict the performance properties of polyhydroxyalkanoates (PHAs), a group of biosourced polyesters featuring excellent performance, to guide future design and synthesis experiments. A deep neural network (DNN) machine learning model was built for predicting the glass transition temperature, Tg, of PHA homo- and copolymers. Molecular fingerprints were used to capture the structural and atomic information of PHA monomers. The other input variables included the molecular weight, the polydispersity index, and the percentage of each monomer in the homo- and copolymers. The results indicate that the DNN model achieves high accuracy in estimation of the glass transition temperature of PHAs. In addition, the symmetry of the DNN model is ensured by incorporating symmetry data in the training process. The DNN model achieved better performance than the support vector machine (SVD), a nonlinear ML model and least absolute shrinkage and selection operator (LASSO), a sparse linear regression model. The relative importance of factors affecting the DNN model prediction were analyzed. Sensitivity of the DNN model, including strategies to deal with missing data, were also investigated. Compared with commonly used machine learning models incorporating quantitative structure–property (QSPR) relationships, it does not require an explicit descriptor selection step but shows a comparable performance. The machine learning model framework can be readily extended to predict other properties.},
doi = {10.3390/ma13245701},
journal = {Materials},
number = 24,
volume = 13,
place = {United States},
year = {Mon Dec 14 00:00:00 EST 2020},
month = {Mon Dec 14 00:00:00 EST 2020}
}
Works referenced in this record:
Future scenarios of global plastic waste generation and disposal
journal, January 2019
- Lebreton, Laurent; Andrady, Anthony
- Palgrave Communications, Vol. 5, Issue 1
Anaerobic degradation of bioplastics: A review
journal, October 2018
- Bátori, Veronika; Åkesson, Dan; Zamani, Akram
- Waste Management, Vol. 80
Deep learning
journal, May 2015
- LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
- Nature, Vol. 521, Issue 7553
Using deep neural network with small dataset to predict material defects
journal, January 2019
- Feng, Shuo; Zhou, Huiyu; Dong, Hongbiao
- Materials & Design, Vol. 162
Microbial Polyhydroxyalkanoates (PHAs): Efficient Replacement of Synthetic Polymers
journal, June 2020
- Muneer, Faizan; Rasul, Ijaz; Azeem, Farrukh
- Journal of Polymers and the Environment, Vol. 28, Issue 9
Polyhydroxyalkanoates: Characteristics, production, recent developments and applications
journal, January 2018
- Raza, Zulfiqar Ali; Abid, Sharjeel; Banat, Ibrahim M.
- International Biodeterioration & Biodegradation, Vol. 126
Survey on Deep Neural Networks in Speech and Vision Systems
journal, December 2020
- Alam, M.; Samad, M. D.; Vidyaratne, L.
- Neurocomputing, Vol. 417
Efficient Machine Learning for Big Data: A Review
journal, September 2015
- Al-Jarrah, Omar Y.; Yoo, Paul D.; Muhaidat, Sami
- Big Data Research, Vol. 2, Issue 3
Artificial neural networks: fundamentals, computing, design, and application
journal, December 2000
- Basheer, I. A.; Hajmeer, M.
- Journal of Microbiological Methods, Vol. 43, Issue 1
Medical Application of Microbial Biopolyesters Polyhydroxyalkanoates
journal, January 2009
- Wu, Qiong; Wang, Yang; Chen, Guo-Qiang
- Artificial Cells, Blood Substitutes, and Biotechnology, Vol. 37, Issue 1
Polyhydroxyalkanoates: biodegradable polymers with a range of applications
journal, January 2007
- Philip, S.; Keshavarz, T.; Roy, I.
- Journal of Chemical Technology & Biotechnology, Vol. 82, Issue 3
Microstructure recognition using convolutional neural networks for prediction of ionic conductivity in ceramics
journal, December 2017
- Kondo, Ruho; Yamakawa, Shunsuke; Masuoka, Yumi
- Acta Materialia, Vol. 141
Degradation Rates of Plastics in the Environment
journal, February 2020
- Chamas, Ali; Moon, Hyunjin; Zheng, Jiajia
- ACS Sustainable Chemistry & Engineering, Vol. 8, Issue 9
Polyhydroxyalkanoates: opening doors for a sustainable future
journal, April 2016
- Li, Zibiao; Yang, Jing; Loh, Xian Jun
- NPG Asia Materials, Vol. 8, Issue 4
Controlling the Shape of Molecular Weight Distributions in Coordination Polymerization and Its Impact on Physical Properties
journal, December 2019
- Sifri, Renee J.; Padilla-Vélez, Omar; Coates, Geoffrey W.
- Journal of the American Chemical Society, Vol. 142, Issue 3
A deep neural network combined with molecular fingerprints (DNN-MF) to develop predictive models for hydroxyl radical rate constants of water contaminants
journal, February 2020
- Zhong, Shifa; Hu, Jiajie; Fan, Xudong
- Journal of Hazardous Materials, Vol. 383
Machine-Learning-Based Predictive Modeling of Glass Transition Temperatures: A Case of Polyhydroxyalkanoate Homopolymers and Copolymers
journal, November 2019
- Pilania, Ghanshyam; Iverson, Carl N.; Lookman, Turab
- Journal of Chemical Information and Modeling, Vol. 59, Issue 12
Polyhydroxyalkanoates: bioplastics with a green agenda
journal, June 2010
- Keshavarz, Tajalli; Roy, Ipsita
- Current Opinion in Microbiology, Vol. 13, Issue 3
SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules
journal, February 1988
- Weininger, David
- Journal of Chemical Information and Modeling, Vol. 28, Issue 1
Current developments on polyhydroxyalkanoates synthesis by using halophiles as a promising cell factory
journal, April 2020
- Mitra, Ruchira; Xu, Tong; Xiang, Hua
- Microbial Cell Factories, Vol. 19, Issue 1
Advances in the Applications of Polyhydroxyalkanoate Nanoparticles for Novel Drug Delivery System
journal, January 2013
- Shrivastav, Anupama; Kim, Hae-Yeong; Kim, Young-Rok
- BioMed Research International, Vol. 2013
Plastic waste inputs from land into the ocean
journal, February 2015
- Jambeck, J. R.; Geyer, R.; Wilcox, C.
- Science, Vol. 347, Issue 6223
Maltodextrin molecular weight distribution influence on the glass transition temperature and viscosity in aqueous solutions
journal, November 2004
- Avaltroni, F.; Bouquerand, P.; Normand, V.
- Carbohydrate Polymers, Vol. 58, Issue 3
Start a Research on Biopolymer Polyhydroxyalkanoate (PHA): A Review
journal, March 2014
- Tan, Giin-Yu; Chen, Chia-Lung; Li, Ling
- Polymers, Vol. 6, Issue 3
Effect of the degree of branching on the glass transition temperature of polyesters
journal, February 2012
- Khalyavina, Anna; Häußler, Liane; Lederer, Albena
- Polymer, Vol. 53, Issue 5
Maltodextrin molecular weight distribution influence on the glass transition temperature and viscosity in aqueous solutions
journal, November 2004
- Avaltroni, F.; Bouquerand, P.; Normand, V.
- Carbohydrate Polymers, Vol. 58, Issue 3
SMILES, a chemical language and information system. 1. Introduction to methodology and encoding rules
journal, February 1988
- Weininger, David
- Journal of Chemical Information and Modeling, Vol. 28, Issue 1