STRUCTURE-ACTIVITY RELATIONSHIP AND CLASSIFICATION OF ANTIMICROBIAL PEPTIDES USING SEQUENCE AND PHYSICOCHEMICAL PROPERTIES
DOI:
https://doi.org/10.65327/bp.v12i3.2529Keywords:
antimicrobial peptides, structure–activity relationship, physicochemical properties, machine learning, XGBoostAbstract
Antimicrobial peptides have been regarded as potential candidates as an alternative to conventional antibiotics; however, their activity is a complex interaction between their sequence composition and physicochemical properties. This study aimed at understanding structure–activity relationships and to build machine-learning models to classify the antimicrobial peptides by combining sequence-derived and molecular descriptors. Conflicting annotations were eliminated, and amino-acid composition, hydrophobicity, hydrophobic moment, charge density, length, molecular weight, and isoelectric point and amino-acid composition were analysed for the relevant features. Antimicrobial peptides were generally shorter and more positively charged than the non-antimicrobial peptides and had a unique residue-composition pattern. Logistic regression revealed that four variables (peptide length, net charge, hydrophobic moment and aromaticity) were important independent predictors of whether a peptide was antimicrobial or not. Four classifiers were tested with sequence-only, physicochemical-only and combined feature sets. XGBoost with combined features had the highest performance with 93.6% accuracy, 0.977 ROC-AUC, 0.975 PR-AUC and 0.871 Matthews correlation coefficient. The discrimination was not so much dependent on peptide size, as evidenced by the good predictive performance of length-controlled sensitivity analysis. It is shown that these results support the usefulness of using interpretable physicochemical descriptors together with sequence data for robust prediction of antimicrobial peptides.
Downloads
References
Al-Omari, A. M., Akkam, Y. H., Zyout, A. A., Younis, S. A., Tawalbeh, S. M., Al-Sawalmeh, K., ... & Arnold, J. (2024). Accelerating antimicrobial peptide design: Leveraging deep learning for rapid discovery. Plos one, 19(12), e0315477.
Brizuela, C. A., Liu, G., Stokes, J. M., & de la Fuente‐Nunez, C. (2025). AI methods for antimicrobial peptides: progress and challenges. Microbial Biotechnology, 18(1), e70072.
Bucataru, C., & Ciobanasu, C. (2024). Antimicrobial peptides: Opportunities and challenges in overcoming resistance. Microbiological research, 286, 127822.
Cardoso, M. H., Orozco, R. Q., Rezende, S. B., Rodrigues, G., Oshiro, K. G., Cândido, E. S., & Franco, O. L. (2020). Computer-aided design of antimicrobial peptides: are we generating effective drug candidates?. Frontiers in microbiology, 10, 3097.
Cresti, L., Cappello, G., & Pini, A. (2024). Antimicrobial peptides towards clinical application—a long history to be concluded. International Journal of Molecular Sciences, 25(9), 4870.
Gagat, P., Ostrówka, M., Duda-Madej, A., & Mackiewicz, P. (2024). Enhancing antimicrobial peptide activity through modifications of charge, hydrophobicity, and structure. International Journal of Molecular Sciences, 25(19), 10821.
Gani, Z., Kumar, A., Raje, M., & Raje, C. I. (2025). Antimicrobial peptides: An alternative strategy to combat antimicrobial resistance. Drug Discovery Today, 30(2), 104305.
Girdhar, M., Sen, A., Nigam, A., Oswalia, J., Kumar, S., & Gupta, R. (2024). Antimicrobial peptide-based strategies to overcome antimicrobial resistance. Archives of Microbiology, 206(10), 411.
Goles, M., Daza, A., Cabas-Mora, G., Sarmiento-Varón, L., Sepúlveda-Yañez, J., Anvari-Kazemabad, H., ... & Medina-Ortiz, D. (2024). Peptide-based drug discovery through artificial intelligence: towards an autonomous design of therapeutic peptides. Briefings in bioinformatics, 25(4), bbae275.
He, S., Stone, T. A., & Deber, C. M. (2021). Uncoupling amphipathicity and hydrophobicity: Role of charge clustering in membrane interactions of cationic antimicrobial peptides. Biochemistry, 60(34), 2586-2592.
Huynh, L., Velásquez, J., Rabara, R., Basu, S., Nguyen, H. B., & Gupta, G. (2021). Rational design of antimicrobial peptides targeting Gram-negative bacteria. Computational Biology and Chemistry, 92, 107475.
Islam, M. M., Asif, F., Zaman, S. U., Arnab, M. K. H., Rahman, M. M., & Hasan, M. (2023). Effect of charge on the antimicrobial activity of alpha-helical amphibian antimicrobial peptide. Current Research in Microbial Sciences, 4, 100182.
Li, C., Sutherland, D., Richter, A., Coombe, L., Yanai, A., Warren, R. L., ... & Birol, I. (2024). De novo synthetic antimicrobial peptide design with a recurrent neural network. Protein Science, 33(8), e5088.
Lin, C., Xiong, S., Cui, F., Zhang, Z., Shi, H., & Wei, L. (2025). Deep learning in antimicrobial peptide prediction. Journal of Chemical Information and Modeling, 65(14), 7373-7392.
Lin, T. T., Yang, L. Y., Lu, I. H., Cheng, W. C., Hsu, Z. R., Chen, S. H., & Lin, C. Y. (2021). AI4AMP: an antimicrobial peptide predictor using physicochemical property-based encoding method and deep learning. Msystems, 6(6), e00299-21.
Malshikare, H., Priyakumar, U. D., Chatterjee, P., & Sengupta, D. (2026). Mechanistic principles of antimicrobial peptides uncovered by charge density-based machine learning. Chemical Communications, 62(13), 4067-4070.
Marczak, B., Bocian, A., & Łyskowski, A. (2026). MarLys AMP database - MLAMP_db (Version 3) [Data set]. Mendeley Data. https://doi.org/10.17632/w4hb5grjwb.3
Mondal, R. K., Sen, D., Arya, A., & Samanta, S. K. (2023). Developing anti-microbial peptide database version 1 to provide comprehensive and exhaustive resource of manually curated AMPs. Scientific Reports, 13(1), 17843.
Pandit, G., Chowdhury, N., Abdul Mohid, S., Bidkar, A. P., Bhunia, A., & Chatterjee, S. (2021). Effect of secondary structure and side chain length of hydrophobic amino acid residues on the antimicrobial activity and toxicity of 14‐residue‐long de novo AMPs. ChemMedChem, 16(2), 355-367.
Pirtskhalava, M., Amstrong, A. A., Grigolava, M., Chubinidze, M., Alimbarashvili, E., Vishnepolsky, B., ... & Tartakovsky, M. (2021). DBAASP v3: database of antimicrobial/cytotoxic activity and structure of peptides as a resource for development of new therapeutics. Nucleic acids research, 49(D1), D288-D297.
Roy, K. (2020). Antimicrobial peptides (Version 1) [Data set]. Mendeley Data. https://doi.org/10.17632/37gbg3p3nr.1
Schifano, N. P., & Caputo, G. A. (2022). Investigation of the role of hydrophobic amino acids on the structure–activity relationship in the antimicrobial venom peptide ponericin L1. The Journal of membrane biology, 255(4), 537-551.
Sun, Z., Wang, X., Kuang, S., Song, J., & de la Fuente-Nunez, C. (2026). From generation to validation: Deep generative models for antimicrobial peptide discovery. Current Opinion in Chemical Biology, 93, 102685.
Szymczak, P., & Szczurek, E. (2023). Artificial intelligence-driven antimicrobial peptide discovery. Current Opinion in Structural Biology, 83, 102733.
Teimouri, H., Medvedeva, A., & Kolomeisky, A. B. (2023). Bacteria-specific feature selection for enhanced antimicrobial peptide activity predictions using machine-learning methods. Journal of Chemical Information and Modeling, 63(6), 1723-1733.
Van der Walt, M., Moller, D. S., Van Wyk, R. J., Ferguson, P. M., Hind, C. K., Clifford, M., ... & Gaspar, A. R. (2024). QSAR reveals decreased lipophilicity of polar residues determines the selectivity of antimicrobial peptide activity. ACS omega, 9(24), 26030-26049.
Wan, F., Wong, F., Collins, J. J., & de la Fuente-Nunez, C. (2024). Machine learning for antimicrobial peptide identification and design. Nature Reviews Bioengineering, 2(5), 392-407.
Wang, G. (2022). Unifying the classification of antimicrobial peptides in the antimicrobial peptide database. In Methods in enzymology (Vol. 663, pp. 1-18). Academic Press.
Wang, G. (2023). The antimicrobial peptide database is 20 years old: Recent developments and future directions. Protein Science, 32(10), e4778.
Wang, G., Vaisman, I. I., & Van Hoek, M. L. (2022). Machine learning prediction of antimicrobial peptides. In Computational peptide science: Methods and protocols (pp. 1-37). New York, NY: Springer US.
Wang, X. F., Tang, J. Y., Sun, J., Dorje, S., Sun, T. Q., Peng, B., ... & Wang, D. B. (2024). ProT‐Diff: a modularized and efficient strategy for de novo generation of antimicrobial peptide sequences by integrating protein language and diffusion models. Advanced Science, 11(43), 2406305.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Dr. Mustafa Alhaji Isa, Dr. Stephen Kayode Ojo, Dr. Tope Tunji Odunitan
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
