Department of Electronics and Information Engineering, Jeonbuk National University, Jeonju-si, 54896 Jeollabuk-do, South Korea.
School of International Engineering and Science, Jeonbuk National University, Jeonju-si, 54896 Jeollabuk-do, South Korea.
J Chem Inf Model. 2024 Jul 8;64(13):4941-4957. doi: 10.1021/acs.jcim.4c00295. Epub 2024 Jun 14.
Anticancer peptides (ACPs) play a vital role in selectively targeting and eliminating cancer cells. Evaluating and comparing predictions from various machine learning (ML) and deep learning (DL) techniques is challenging but crucial for anticancer drug research. We conducted a comprehensive analysis of 15 ML and 10 DL models, including the models released after 2022, and found that support vector machines (SVMs) with feature combination and selection significantly enhance overall performance. DL models, especially convolutional neural networks (CNNs) with light gradient boosting machine (LGBM) based feature selection approaches, demonstrate improved characterization. Assessment using a new test data set (ACP10) identifies ACPred, MLACP 2.0, AI4ACP, mACPred, and AntiCP2.0_AAC as successive optimal predictors, showcasing robust performance. Our review underscores current prediction tool limitations and advocates for an omnidirectional ACP prediction framework to propel ongoing research.
抗癌肽 (ACPs) 在选择性靶向和消除癌细胞方面发挥着重要作用。评估和比较来自各种机器学习 (ML) 和深度学习 (DL) 技术的预测结果具有挑战性，但对于抗癌药物研究至关重要。我们对 15 种 ML 和 10 种 DL 模型进行了全面分析，包括 2022 年后发布的模型，发现通过特征组合和选择的支持向量机 (SVM) 可显著提高整体性能。DL 模型，特别是基于轻梯度提升机 (LGBM) 的特征选择方法的卷积神经网络 (CNN)，表现出更好的特征描述能力。使用新测试数据集 (ACP10) 的评估结果表明，ACPred、MLACP 2.0、AI4ACP、mACPred 和 AntiCP2.0_AAC 是连续的最佳预测器，展示了稳健的性能。我们的综述强调了当前预测工具的局限性，并倡导建立一个全方位的 ACP 预测框架，以推动正在进行的研究。
