Sequence-based peptide identification, generation, and property prediction with deep learning: a review

Over the past few years, deep learning has demonstrated itself to be a powerful tool in many areas, especially bioinformatics. With its previous success in DNA and protein related studies, deep learning has now been brought to the field of peptide science as well. It has been widely used in sequence...

Full description

Saved in:
Bibliographic Details
Published inMolecular systems design & engineering Vol. 6; no. 6; pp. 46 - 428
Main Authors Chen, Xumin, Li, Chen, Bernards, Matthew T, Shi, Yao, Shao, Qing, He, Yi
Format Journal Article
LanguageEnglish
Published Cambridge Royal Society of Chemistry 08.06.2021
Subjects
Algorithms
Antiinfectives and antibacterials
Artificial neural networks
Bioinformatics
Deep learning
Generative adversarial networks
Machine learning
Neural networks
Optimization
Peptides
Recurrent neural networks
Workflow
Online AccessGet full text

Cover

More Information
Summary:Over the past few years, deep learning has demonstrated itself to be a powerful tool in many areas, especially bioinformatics. With its previous success in DNA and protein related studies, deep learning has now been brought to the field of peptide science as well. It has been widely used in sequence-based peptide identification, generation, and property prediction. The publications on this subject over the past two years are summarized in this review. The deep learning models reported are mainly convolutional neural networks, recurrent neural networks, hybrid models, transformers, and other generative models like variational autoencoders and generative adversarial networks, as well as algorithms like input optimization. Application areas include antimicrobial peptides, signal peptides, and major histocompatibility complex binding peptides, among others. This review develops content according to the general workflow of deep learning, while illustrating adaptations and techniques specific to certain example problems. Some issues and future directions are also discussed, such as approaches for model interpretation, benchmark datasets, automation in deep learning, and rational peptide design techniques. This article reviews recent work that uses deep learning algorithms to identify and generate functional peptides as well as predict their biological properties.
Bibliography:Dr. Matthew Bernards is an associate professor in the Department of Chemical and Biological Engineering at the University of Idaho. He also serves as the Director of the NASA Idaho Space Grant Consortium and Idaho NASA EPSCoR programs. He graduated with his Ph.D. in Chemical Engineering from the University of Washington in 2008. The Bernards research group is focused on multiple aspects of materials science and engineering including computational and experimental investigations into the interactions that occur between biological entities and material interfaces.
peptide design.
Dr. Qing Shao obtained his Ph.D. with Dr. Shaoyi Jiang in Chemical Engineering at the University of Washington in 2014, then did postdoctoral research with Dr. Carol Hall at North Carolina State University. He started as an assistant professor in the chemical and materials engineering department at the University of Kentucky in 2018. His current research focuses on understanding and designing solvents for energy, environmental and biological applications using computational approaches.
de novo
Mr. Xumin Chen received his bachelor's degree in chemical engineering from Zhejiang University in 2019. He is a postgraduate in the College of Chemical and Biological Engineering at Zhejiang University. His current work focuses on biomaterials development, high-throughput peptide screening with machine learning or deep learning algorithms, and
Mr. Chen Li received his M.S. degree in chemical engineering from China University of Petroleum (Beijing) in 2020. He worked at the Institute of Process Engineering, Chinese Academy of Sciences between 2018-2020, dedicated to the molecular dynamics simulation of dielectrics and biomolecules. In August 2020, he moved to work as a research assistant at the Department of Polymer Science and Engineering of Zhejiang University. His current work focuses on high-throughput screening of peptides, including machine learning algorithms and molecular dynamics simulations.
Dr. Yao Shi is a professor in the College of Chemical and Biological Engineering at Zhejiang University. He obtained his Ph.D. in Chemical Engineering at Zhejiang University in 1991. He was a senior visiting scholar at the University of California, Berkeley and Lawrence Berkeley National Laboratory. His current research interests include developing new technologies that protect human health and promote sustainability.
Dr. Yi He is an associate professor in the College of Chemical and Biological Engineering at Zhejiang University. He is also an affiliate associate professor in the Department of Chemical Engineering at the University of Washington, Seattle. He serves as a member of the editorial board of Molecular Simulation. He graduated with his Ph.D. in Chemical Engineering from the University of Washington in 2008. His current work focuses on molecular simulations and machine learning for biomaterials, biomolecules, and environmental engineering.
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2058-9689
2058-9689
DOI:10.1039/d0me00161a