ITP-Pred: an interpretable method for predicting, therapeutic peptides with fused features low-dimension representation

被引:65
作者
Cai, Lijun [1 ]
Wang, Li [1 ]
Fu, Xiangzheng [1 ]
Xia, Chenxing [2 ]
Zeng, Xiangxiang [1 ]
Zou, Quan [3 ]
机构
[1] Hunan Univ, Changsha, Hunan, Peoples R China
[2] Anhui Univ Sci & Technol, Huainan, Anhui, Peoples R China
[3] Univ Elect Sci & Technol China, Chengdu, Sichuan, Peoples R China
基金
中国国家自然科学基金; 中国博士后科学基金;
关键词
therapeutic peptides prediction; CNN-BiLSTM; feature fusion; interpretability analysis; CELL-PENETRATING PEPTIDES; SECONDARY STRUCTURE; CHEMICAL SPACE; PROTEIN; DELIVERY; CLASSIFICATION; IDENTIFICATION;
D O I
10.1093/bib/bbaa367
中图分类号
Q5 [生物化学];
学科分类号
071010 ; 081704 ;
摘要
The peptide therapeutics market is providing new opportunities for the biotechnology and pharmaceutical industries. Therefore, identifying therapeutic peptides and exploring their properties are important. Although several studies have proposed different machine learning methods to predict peptides as being therapeutic peptides, most do not explain the decision factors of model in detail. In this work, an Interpretable Therapeutic Peptide Prediction (ITP-Pred) model based on efficient feature fusion was developed. First, we proposed three kinds of feature descriptors based on sequence and physicochemical property encoded, namely amino acid composition (AAC), group AAC and coding autocorrelation, and concatenated them to obtain the feature representation of therapeutic peptide. Then, we input it into the CNN-Bi-directional Long Short-Term Memory (BiLSTM) model to automatically learn recognition of therapeutic peptides. The cross-validation and independent verification experiments results indicated that ITP-Pred has a higher prediction performance on the benchmark dataset than other comparison methods. Finally, we analyzed the output of the model from two aspects: sequence order and physical and chemical properties, mining important features as guidance for the design of better models that can complement existing methods.
引用
收藏
页数:12
相关论文
共 47 条
[31]  
PALAU J, 1982, INT J PEPT PROT RES, V19, P394
[32]  
Presente A, 2013, CURR PHARM DESIGN, V19, P2943
[33]   Cell penetrating peptides: a comparative transport analysis for 474 sequence motifs [J].
Ramaker, Katrin ;
Henkel, Maik ;
Krause, Thorsten ;
Roeckendorf, Niels ;
Frey, Andreas .
DRUG DELIVERY, 2018, 25 (01) :928-937
[34]   AntiAngioPred: A Server for Prediction of Anti-Angiogenic Peptides [J].
Ramaprasad, Azhagiya Singam Ettayapuram ;
Singh, Sandeep ;
Gajendra, Raghava P. S. ;
Venkatesan, Subramanian .
PLOS ONE, 2015, 10 (09)
[35]  
Saha Sudipto, 2007, In Silico Biology, V7, P369
[36]   Identification of Short Hydrophobic Cell-Penetrating Peptides for Cytosolic Peptide Delivery by Rational Design [J].
Schmid, Samuel ;
Adjobo-Hermans, Merel J. W. ;
Kohze, Robin ;
Enderle, Thilo ;
Brock, Roland ;
Milletti, Francesca .
BIOCONJUGATE CHEMISTRY, 2017, 28 (02) :382-389
[37]   Enhancing cellular uptake of activable cell-penetrating peptide-doxorubicin conjugate by enzymatic cleavage [J].
Shi, Nian-Qiu ;
Gao, Wei ;
Xiang, Bai ;
Qi, Xian-Rong .
INTERNATIONAL JOURNAL OF NANOMEDICINE, 2012, 7 :1613-1621
[38]   Peptide Design Principles for Antimicrobial Applications [J].
Torres, Marcelo D. T. ;
Sothiselvam, Shanmugapriya ;
Lu, Timothy K. ;
de la Fuente-Nunez, Cesar .
JOURNAL OF MOLECULAR BIOLOGY, 2019, 431 (18) :3547-3567
[39]  
UniProt Consortium, 2019, Nucleic Acids Res., V47, pD506, DOI [DOI 10.1093/nar/gky1049, 10.1093/nar/gky1049]
[40]   Comparative analysis and prediction of quorum-sensing peptides using feature representation learning and machine learning algorithms [J].
Wei, Leyi ;
Hu, Jie ;
Li, Fuyi ;
Song, Jiangning ;
Su, Ran ;
Zou, Quan .
BRIEFINGS IN BIOINFORMATICS, 2020, 21 (01) :106-119