iAMPM: A Deep Learning Approach for Improved Recognition of Antimicrobial Peptides
Abstract
by swift killing capabilities, low toxicity profiles, and broad-spectrum activity. These attributes render AMPs as promising candidates for addressing the pressing issue of escalating antibiotic resistance. In comparison to the labor-intensive wet-laboratory techniques and machine
learning approaches for antimicrobial peptide identification, deep learning-based methodologies offer a more streamlined and convenient
alternative that obviates the need for intricate feature engineering processes. This study presents a novel deep learning framework, designated
as iAMPM, for the identification of antimicrobial peptides. This model makes predictions by extracting the spatial structural features of peptides in the global dimension. A graph neural network module, comprising a MixHop layer, a global average pooling layer, and a self-attentive
module, is employed for the extraction of peptide spatial information. The three-dimensional peptide structure was initially calculated with
atomic precision using the AlphaFold protein structure prediction module. Thereafter, the peptide residue contact maps were generated
computationally based on this peptide structure. Subsequently, the pre-trained model (ESM) is employed to extract peptide residue features,
thereby generating peptide residue contact map embeddings. The data utilized in this experiment were obtained from a number of databases,
including dbAMP, APD3, CAMP, DRAMP, and UniProt. In the independent dataset, the Sn metrics and AUC values of iAMPM demonstrated
superior performance compared to the best-performing sequence-based iAMPCN, with an improvement of 9.08% and 4.2%, respectively.
Keywords
Full Text:
PDFReferences
[1] DCosta VM, King CE, Kalan L, et al. Antibiotic resistance is ancient. Nature 2011; 477(7365): 457-461.
[2] Erdem Buyukkiraz M, Kesmen Z. Antimicrobial peptides (AMPs): A promising class of antimicrobial compounds. Journal of applied
microbiology 2022; 132(3): 1573-1596.
[3] Zhang L, Gallo RL. Antimicrobial peptides. Current biology 2016; 26(1): R14-R19.
[4] Bahar AA, Ren D. Antimicrobial peptides. Pharmaceuticals 2013; 6(12): 1543-1575.
[5] Magana M, Pushpanathan M, Santos AL, et al. The value of antimicrobial peptides in the age of resistance. The lancet infectious diseases 2020; 20(9): e216-e230.
[6] Veltri D, Kamath U, Shehu A. Deep learning improves antimicrobial peptide recognition. Bioinformatics 2018; 34(16): 2740-2747.
[7] Su X, Xu J, Yin Y, Quan X, Zhang H. Antimicrobial peptide identification using multi-scale convolutional network. BMC bioinformatics
2019; 20: 1-10.
[8] Xu J, Li F, Li C, et al. iAMPCN: a deep-learning approach for identifying antimicrobial peptides and their functional activities. Briefings
in Bioinformatics 2023; 24(4): bbad240.
[9] Jumper J, Evans R, Pritzel A, et al. Highly accurate protein structure prediction with AlphaFold. nature 2021; 596(7873): 583-589.
[10] Lin Z, Akin H, Rao R, et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science 2023;
379(6637): 1123-1130.
[11] Abu-El-Haija S, Perozzi B, Kapoor A, et al. Mixhop: Higher-order graph convolutional architectures via sparsified neighborhood mixing. In: international conference on machine learning; 2019. pp. 21-29
[12] Vaswani A. Attention is all you need. arXiv preprint arXiv:1706.03762 2017.
[13] Chicco D, Jurman G. The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification
evaluation. BMC genomics 2020; 21: 1-13.
[14] Jhong JH, Yao L, Pang Y, et al. dbAMP 2.0: updated resource for antimicrobial peptides with an enhanced scanning method for genomic
and proteomic data. Nucleic acids research 2022; 50(D1): D460-D470.
[15] Wang G, Li X, Wang Z. APD3: the antimicrobial peptide database as a tool for research and education. Nucleic acids research 2016;
44(D1): D1087-D1093.
[16] UniProt: the universal protein knowledgebase in 2023. Nucleic acids research 2023; 51(D1): D523-D531.
[17] Gull S, Shamim N, Minhas F. AMAP: Hierarchical multi-label prediction of biologically active and antimicrobial peptides. Computers
in biology and medicine 2019; 107: 172-181.
[18] Chung CR, Kuo TR, Wu LC, Lee TY, Horng JT. Characterization and identification of antimicrobial peptides with different functional
activities. Briefings in bioinformatics 2020; 21(3): 1098-1114.
[19] Xiao X, Shao YT, Cheng X, Stamatovic B. iAMP-CA2L: a new CNN-BiLSTM-SVM classifier based on cellular automata image for
identifying antimicrobial peptides and their functional types. Briefings in bioinformatics 2021; 22(6): bbab209.
DOI: http://dx.doi.org/10.70711/aitr.v2i6.5734
Refbacks
- There are currently no refbacks.
