Abstract
Anticancer peptides (ACPs) hold considerable promise
as therapeutic agents, but screening candidate peptides in the laboratory is
both expensive and slow, so reliable computational tools are in demand. A
number of machine-learning and deep-learning predictors already exist, yet
their accuracy on small benchmark datasets is still modest, and the benefit of
merging different sequence descriptors has not been fully exploited. This paper
describes an ensemble-learning workflow for identifying ACPs that draws on
multiple complementary views of each sequence. Each peptide is first
represented by amino acid composition (AAC) and composition of k-spaced amino
acid pairs (CKSAAP), which jointly capture overall composition and local order;
two physicochemical attributes (average charge and average hydrophobicity) are
appended to these features. The resulting 822 dimensions are Z-score
standardized and then pruned to 262 informative features by an XGBoost-based
selector. Prediction is made by a soft-voting ensemble that pools a Random
Forest and a tuned XGBoost classifier. Evaluated on the ACP-2.0 benchmark, this
ensemble reaches 77.5% accuracy, a Matthews correlation coefficient of 0.551,
and an area under the ROC curve (AUC) of 0.850, beating both base learners and
the AntiCP 2.0 baseline while staying competitive with deep-learning methods
that are far more demanding computationally. The results point to a practical,
accurate tool for large-scale ACP screening.
References
[1] Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2022: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229-63.
[2] Ghadiri N, Zali H, Vafaei M, et al. Bioactive peptides: An alternative therapeutic approach for cancer management. Front Immunol. 2024;15:1310443.
[3] Xu M, Li W, Wang X, et al. Integrating traditional machine learning and deep learning for precision screening of anticancer peptides: A novel approach for efficient drug discovery. ACS Omega. 2024;9(14):16820-31.
[4] Kim H, Kim J, Lee I. Artificial intelligence in drug discovery: A comprehensive review of data-driven and machine learning approaches. Biotechnol Bioprocess Eng. 2020;25(6):895-930.
[5] Schaduangrat N, Lohavanichbutr P, Saeaeung A, et al. Acpred: A computational tool for the prediction and analysis of anticancer peptides. Molecules. 2019;24(10):1973.
[6] Ahmed S, Mugal I, Ijaz MT, et al. Acp-Mhcnn: An accurate multi-headed deep-convolutional neural network to predict anticancer peptides. Sci Rep. 2021;11:23676.
[7] Yi HC, You ZH, Huang DS, et al. Acp-Dl: A deep learning long short-term memory model to predict anticancer peptides using high-efficiency feature representation. Mol Ther Nucleic Acids. 2019;17:1-9.
[8] Chiangjong W, Chutipongtanate S, Hongeng S. Anticancer peptide: Physicochemical property, functional aspect and trend in clinical application. Int J Oncol. 2020;57(3):678-96.
[9] Van Hilten A, Chevyrev V, Maticzka D, et al. Designing interpretable deep learning applications for functional genomics: A quantitative analysis. Brief Bioinform. 2024;25(5):bbae344.
[10] Yuan Q, Chen J, Zhao H, et al. Prediction of anticancer peptides based on an ensemble model of deep learning and machine learning using ordinal positional encoding. Brief Bioinform. 2023;24(1):bbac630.
[11] Agrawal P, Bhalla S, Chaudhary K, et al. Anticp 2.0: An updated model for predicting anticancer peptides. Brief Bioinform. 2021;22(3):bbaa411.
[12] Zheng B, Chen J, Zhu X, et al. Iacp-Sei: An anticancer peptide identification method incorporating sequence evolutionary information. J Data Sci Intell Syst. 2025;3(1):45-58.
[13] Huang G, Liu X, Li Y, et al. Acp-Dpe: A dual-channel deep learning model for anticancer peptide prediction. IET Syst Biol. 2025;19(1):e70010.
[14] Yao L, You ZH, Huang DS. Acp-Capspred: An explainable computational framework for identification and functional prediction of anticancer peptides based on capsule network. Brief Bioinform. 2024;25(5):bbae380.
How to cite this paper
Ensemble-learning-based Predictor for Anticancer Peptide Identification Using Multi-view Sequence Features
How to cite this paper: Xiaoqin Fan, Xin Zhang, Mingxing Xu, Kexu Wu. (2026) Ensemble-learning-based Predictor for Anticancer Peptide Identification Using Multi-view Sequence Features. Advances in Computer and Communication, 7(3), 187-191.
DOI: http://dx.doi.org/10.26855/acc.2026.09.015