AI4AMP: An antimicrobial peptide predictor using physicochemical property-based encoding method and deep learning

Tzu Tang Lin; Li Yen Yang; I. Hsuan Lu; Wen Chih Cheng; Zhe Ren Hsu; Shu Hwa Chen; Chung Yen Lin

doi:10.1128/mSystems.00299-21

AI4AMP: An antimicrobial peptide predictor using physicochemical property-based encoding method and deep learning

Tzu Tang Lin, Li Yen Yang, I. Hsuan Lu, Wen Chih Cheng, Zhe Ren Hsu, Shu Hwa Chen, Chung Yen Lin

TMU Research Center of Cancer Translational Medicine

Research output: Contribution to journal › Article › peer-review

20 Citations (Scopus)

Abstract

Antimicrobial peptides (AMPs) are innate immune components that have recently stimulated considerable interest among drug developers due to their potential as antibiotic substitutes. AMPs are notable for their fundamental properties of microbial membrane structural interference and the biomedical applications of killing or suppressing microbes. New AMP candidates must be developed to oppose antibiotic resistance. However, the discovery of novel AMPs through wet-lab screening approaches is inefficient and expensive. The prediction model investigated in this study may help accelerate this process. We collected both the up-to-date AMP data set and unbiased negatives based on which the protein-encoding methods and deep learning model for AMPs were investigated. The external testing results indicated that our trained model achieved 90% precision, outperforming current methods. We implemented our model on a user-friendly web server, AI4AMP, to accurately predict the antimicrobial potential of a given protein sequence and perform proteome screening.

Original language	English
Article number	e00299-21
Journal	mSystems
Volume	6
Issue number	6
DOIs	https://doi.org/10.1128/mSystems.00299-21
Publication status	Published - Dec 2021

Keywords

Antimicrobial peptide
Deep learning
Protein-encoding method
Real-world data
Web service

ASJC Scopus subject areas

Microbiology
Ecology, Evolution, Behavior and Systematics
Biochemistry
Physiology
Modelling and Simulation
Molecular Biology
Genetics
Computer Science Applications

Access to Document

10.1128/mSystems.00299-21

Cite this

@article{6fdddbe1d39f40d6bc31cee46cea2833,

title = "AI4AMP: An antimicrobial peptide predictor using physicochemical property-based encoding method and deep learning",

abstract = "Antimicrobial peptides (AMPs) are innate immune components that have recently stimulated considerable interest among drug developers due to their potential as antibiotic substitutes. AMPs are notable for their fundamental properties of microbial membrane structural interference and the biomedical applications of killing or suppressing microbes. New AMP candidates must be developed to oppose antibiotic resistance. However, the discovery of novel AMPs through wet-lab screening approaches is inefficient and expensive. The prediction model investigated in this study may help accelerate this process. We collected both the up-to-date AMP data set and unbiased negatives based on which the protein-encoding methods and deep learning model for AMPs were investigated. The external testing results indicated that our trained model achieved 90% precision, outperforming current methods. We implemented our model on a user-friendly web server, AI4AMP, to accurately predict the antimicrobial potential of a given protein sequence and perform proteome screening.",

keywords = "Antimicrobial peptide, Deep learning, Protein-encoding method, Real-world data, Web service",

author = "Lin, {Tzu Tang} and Yang, {Li Yen} and Lu, {I. Hsuan} and Cheng, {Wen Chih} and Hsu, {Zhe Ren} and Chen, {Shu Hwa} and Lin, {Chung Yen}",

note = "Funding Information: We thank the Ministry of Science and Technology (MOST), Taiwan, and Academia Sinica, Taiwan, for supporting this research and publication through MOST 108-2314-B-001 -002, MOST 108-2321-B-038 -003, and Grand Challenge Seed Program (206d-1090107), respectively. We also thank the Council of Agriculture, Executive Yuan, Taiwan (109AgriS-12.3.1-S-a2). No conflicts of interest are declared. Funding Information: We thank the Ministry of Science and Technology (MOST), Taiwan, and Academia Sinica, Taiwan, for supporting this research and publication through MOST 108-2314-B-001-002, MOST 108-2321-B-038-003, and Grand Challenge Seed Program (206d-1090107), respectively. We also thank the Council of Agriculture, Executive Yuan, Taiwan (109AgriS-12.3.1-S-a2). Publisher Copyright: Copyright {\textcopyright} 2021 Lin et al.",

year = "2021",

month = dec,

doi = "10.1128/mSystems.00299-21",

language = "English",

volume = "6",

journal = "mSystems",

issn = "2379-5077",

publisher = "American Society for Microbiology",

number = "6",

}

TY - JOUR

T1 - AI4AMP

T2 - An antimicrobial peptide predictor using physicochemical property-based encoding method and deep learning

AU - Lin, Tzu Tang

AU - Yang, Li Yen

AU - Lu, I. Hsuan

AU - Cheng, Wen Chih

AU - Hsu, Zhe Ren

AU - Chen, Shu Hwa

AU - Lin, Chung Yen

N1 - Funding Information: We thank the Ministry of Science and Technology (MOST), Taiwan, and Academia Sinica, Taiwan, for supporting this research and publication through MOST 108-2314-B-001 -002, MOST 108-2321-B-038 -003, and Grand Challenge Seed Program (206d-1090107), respectively. We also thank the Council of Agriculture, Executive Yuan, Taiwan (109AgriS-12.3.1-S-a2). No conflicts of interest are declared. Funding Information: We thank the Ministry of Science and Technology (MOST), Taiwan, and Academia Sinica, Taiwan, for supporting this research and publication through MOST 108-2314-B-001-002, MOST 108-2321-B-038-003, and Grand Challenge Seed Program (206d-1090107), respectively. We also thank the Council of Agriculture, Executive Yuan, Taiwan (109AgriS-12.3.1-S-a2). Publisher Copyright: Copyright © 2021 Lin et al.

PY - 2021/12

Y1 - 2021/12

N2 - Antimicrobial peptides (AMPs) are innate immune components that have recently stimulated considerable interest among drug developers due to their potential as antibiotic substitutes. AMPs are notable for their fundamental properties of microbial membrane structural interference and the biomedical applications of killing or suppressing microbes. New AMP candidates must be developed to oppose antibiotic resistance. However, the discovery of novel AMPs through wet-lab screening approaches is inefficient and expensive. The prediction model investigated in this study may help accelerate this process. We collected both the up-to-date AMP data set and unbiased negatives based on which the protein-encoding methods and deep learning model for AMPs were investigated. The external testing results indicated that our trained model achieved 90% precision, outperforming current methods. We implemented our model on a user-friendly web server, AI4AMP, to accurately predict the antimicrobial potential of a given protein sequence and perform proteome screening.

AB - Antimicrobial peptides (AMPs) are innate immune components that have recently stimulated considerable interest among drug developers due to their potential as antibiotic substitutes. AMPs are notable for their fundamental properties of microbial membrane structural interference and the biomedical applications of killing or suppressing microbes. New AMP candidates must be developed to oppose antibiotic resistance. However, the discovery of novel AMPs through wet-lab screening approaches is inefficient and expensive. The prediction model investigated in this study may help accelerate this process. We collected both the up-to-date AMP data set and unbiased negatives based on which the protein-encoding methods and deep learning model for AMPs were investigated. The external testing results indicated that our trained model achieved 90% precision, outperforming current methods. We implemented our model on a user-friendly web server, AI4AMP, to accurately predict the antimicrobial potential of a given protein sequence and perform proteome screening.

KW - Antimicrobial peptide

KW - Deep learning

KW - Protein-encoding method

KW - Real-world data

KW - Web service

UR - http://www.scopus.com/inward/record.url?scp=85120373864&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85120373864&partnerID=8YFLogxK

U2 - 10.1128/mSystems.00299-21

DO - 10.1128/mSystems.00299-21

M3 - Article

AN - SCOPUS:85120373864

SN - 2379-5077

VL - 6

JO - mSystems

JF - mSystems

IS - 6

M1 - e00299-21

ER -

AI4AMP: An antimicrobial peptide predictor using physicochemical property-based encoding method and deep learning

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this