| 引用本文: | 李尧尧,李钧翔,李旭辉,张瑾.人工智能在细胞穿透肽预测中的研究进展[J].生物信息学,2026,24(1):14-22. |
| LI Yaoyao,LI Junxiang,LI Xuhui,ZHANG Jin.Research progress of artificial intelligence in cell penetrating peptide prediction[J].Chinese Journal of Bioinformatics,2026,24(1):14-22. |
|
| 本文已被：浏览 52次 下载 28次 |
码上扫一扫！ |
|
|
| 人工智能在细胞穿透肽预测中的研究进展 |
|
李尧尧1,2, 李钧翔3,4, 李旭辉5, 张瑾2
|
|
(1. 浙江理工大学 生命科学与医药学院,杭州 310018;2. 嘉兴大学 生物与化学工程学院,浙江 嘉兴314000;3. 浙江清华长三角研究院 衰老科学创新研发中心,浙江 嘉兴341001;4. 禾美生物科技(浙江)有限公司,浙江 嘉兴341001;5. 浙江省应用酶学重点实验室(浙江清华长三角研究院),浙江 嘉兴 314006)
|
|
| 摘要: |
| 细胞穿透肽(Cell penetrating peptides, CPPs)是指能通过直接转运或内吞作用进入细胞的多肽,一般不超过30个氨基酸。CPPs可携带多种活性物质进入细胞,有望成为新型药物的递送载体。传统实验方法获得CPPs具有工作量大、通量低、周期长等问题；随着计算生物学的发展,基于机器学习算法的人工智能模型提高了候选CPPs的预测效率。本文介绍了基于支持向量机、随机森林、极限学习机、极端随机树和其他深度学习的CPPs预测方法,最后探讨了人工智能预测CPPs存在的挑战和对未来的展望。本文旨在为从事CPPs相关研究的学者利用计算生物学工具预测和初步筛选CPPs提供基础和泛化的理论指导。 |
| 关键词: 人工智能 细胞穿透肽 CPP预测 机器学习 |
| DOI：10.12113/202409008 |
| 分类号:R978.1+6 |
| 文献标识码:A |
| 基金项目:国家自然科学基金项目(No.32172708)；浙江省自然基金重点项目(No.LZ23C170002). |
|
| Research progress of artificial intelligence in cell penetrating peptide prediction |
|
LI Yaoyao1,2, LI Junxiang3,4, LI Xuhui5, ZHANG Jin2
|
|
(1. School of Life Sciences and Medicine, Zhejiang Sci-Tech University, Hangzhou 310018, China;2. School of Biological and Chemical Engineering, Jiaxing University, Jianxing 314001, Zhejiang, China;3. AGECODE R&D CENTER, Yangtze Delta Region Institute of Tsinghua University, Jianxing 314001, Zhejiang, China;4. Hemei Biotechnology (Zhejiang) Co., Ltd., Jianxing 314001, Zhejiang, China;5. Key Laboratory of Applied Enzymology, Zhejiang Province(Yangtze delta region institute of Tsinghua University), Jianxing 314001, Zhejiang, China)
|
| Abstract: |
| Cell penetrating peptides(CPPs) refer to polypeptides that can enter cells through direct transport or endocytosis, and generally do not exceed 30 amino acids. CPPs can carry a variety of active substances into cells and is expected to become a new drug delivery carrier. The traditional experimental method to obtain CPPs has many problems, such as heavy workload, low flux and long cycle. With the development of computational biology, the artificial intelligence model based on machine learningalgorithm improves the prediction efficiency of candidate CPPs. This paper introduces the prediction method of CPPs based on support vector machine, random forest, extreme learning machine, extreme random tree and deep learning, and discusses the influence of sequence feature extraction and insufficient training set on the accuracy of artificial intelligence prediction of CPPs. It is believed that with the development of artificial intelligence technology, researchers will be able to develop a CPPs prediction model with higher accuracy and stronger generalization ability. |
| Key words: Artificial intelligence Cell penetrating peptide CPPs prediction Machine learning |
