biostudies-literatureLi XChongqing Talent Program Projectthe Technology Innovation and Application Development Key Project of Chongqingthe National Natural Science Foundation of China4143https://www.ebi.ac.uk/biostudies/studies/S-EPMC12692575biostudies-literatureUnknown14(23)Accurate prediction of hemolytic peptides is essential for peptide safety evaluation and therapeutic design; however, existing models remain constrained by limited accuracy and interpretability. To overcome these challenges, we propose a regression framework that integrates embeddings from a protein language model with handcrafted amino acid descriptors. Specifically, sequence representations derived from the ESM2_t33 model are fused with physicochemical amino acid descriptor features, and key predictive variables are selected through a three-stage strategy involving variance filtering, F-test ranking, and mutual information analysis. The final ensemble model, composed of Random Forest, Extremely Randomized Trees, Gradient Boosting, eXtreme Gradient Boosting (XGBoost), and Ridge Regression, achieved a coefficient of determination (R²) of 0.57 and a correlation coefficient (R) of 0.76 on the test set, outperforming previous approaches. To enhance interpretability, we applied Shapley value analysis and the Calibrated_Explanation algorithm to quantify feature contributions and generate reliable sample-specific explanations. The trained model has been deployed online as HemPepPred, a tool for predicting hemolytic concentration (HC₅₀) values, which provides a practical platform for rational peptide design and safety assessment.Foods (Basel, Switzerland)HemPepPred: Quantitative Prediction of Peptide Hemolytic Activity Based on Machine Learning and Protein Language Model-Derived Features.PMC12692575cstc2024ycjhbgzxm0113CSTB2024TIAD-KPX001432172196Li XZhuo XLiang GLiang XYu SZhao WfalseHemPepPred: Quantitative Prediction of Peptide Hemolytic Activity Based on Machine Learning and Protein Language Model-Derived Features.Accurate prediction of hemolytic peptides is essential for peptide safety evaluation and therapeutic design; however, existing models remain constrained by limited accuracy and interpretability. To overcome these challenges, we propose a regression framework that integrates embeddings from a protein language model with handcrafted amino acid descriptors. Specifically, sequence representations derived from the ESM2_t33 model are fused with physicochemical amino acid descriptor features, and key predictive variables are selected through a three-stage strategy involving variance filtering, F-test ranking, and mutual information analysis. The final ensemble model, composed of Random Forest, Extremely Randomized Trees, Gradient Boosting, eXtreme Gradient Boosting (XGBoost), and Ridge Regression, achieved a coefficient of determination (R²) of 0.57 and a correlation coefficient (R) of 0.76 on the test set, outperforming previous approaches. To enhance interpretability, we applied Shapley value analysis and the Calibrated_Explanation algorithm to quantify feature contributions and generate reliable sample-specific explanations. The trained model has been deployed online as HemPepPred, a tool for predicting hemolytic concentration (HC₅₀) values, which provides a practical platform for rational peptide design and safety assessment.2025-01-01T00:00:00Z2025 Dec2026-05-26T14:07:08.214Z2026-05-24T03:11:07.81ZS-EPMC126925754137608010.3390/foods14234143