计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240800024-6.doi: 10.11896/jsjkx.240800024
冉琴, 阮小利, 徐婧, 李少波, 胡丙齐
RAN Qin, RUAN Xiaoli, XU Jing, LI Shaobo, HU Bingqi
摘要： 在生物医学领域,治疗肽作为传统抗生素药物的有效替代品,因其低毒性、高吸收率和高生物活性而被广泛应用于疾病治疗。然而,目前从深度学习的角度预测肽功能的研究还仍有较大改进空间。因此,基于公开的多功能治疗肽数据集,提出了一种基于投影剃度下降的多编码神经网络(PrMFTP-PGD)。首先,结合了多头注意力机制的多编码器提取输入向量的特征并获得较好的表示能力。然后,引入线性注意力机制进一步增强对特征的表示和提取能力。最后,通过投影梯度下降的对抗训练缓解多功能治疗肽数据集中固有的类不平衡问题。在独立测试集上与MPMAB,MLBP,PrMFTP,SP-RNN和ETFC方法进行比较,在精确率、覆盖率、准确率和绝对正确率指标中最大分别提升了2.55%,2.81%,2.59%和2.39%,结果表明,所提方法能够增强模型捕捉序列特征的能力,以更好地对多功能治疗肽进行预测。
中图分类号:
| [1]HUANG J R,GENG Y L G,GU Y M,et al.Research Status of Antibiotic Abuse[J].Advances in Clinical Medicine,2023,13(9):14311-14317. [2]TANG W,DAI R,YAN W,et al.Identifying multi-functionalbioactive peptide functions using multi-label deep learning[J].Briefings in Bioinformatics,2022,23(1):bbab414. [3]YAN K,LV H,WEN J,et al.PreTP-Stack:Prediction of Therapeutic Peptides Based on the Stacked Ensemble Learing[J].IEEE/ACM Transactions on Computational Biology and Bioinformatics,2023,20(2):1337-1344. [4]YAN W,TANG W,WANG L,et al.PrMFTP:Multi-functional therapeutic peptides prediction based on multi-head self-attention mechanism and class weight optimization[J].PLOS Computational Biology,2022,18(9):e1010511. [5]PORTO W F,PIRES A S,FRANCO O L.Computational tools for exploring sequence databases as a resource for antimicrobial peptides[J].Biotechnology Advances,2017,35(3):337-349. [6]MENG C,HU Y,ZHANG Y,et al.PSBP-SVM:A MachineLearning-Based Computational Identifier for Predicting Polystyrene Binding Peptides[J].Frontiers in Bioengineering and Biotechnology,2020,8:245. [7]ZHANG J,ZHANG Z,PU L,et al.AIEpred:An Ensemble Predictive Model of Classifier Chain to Identify Anti-Inflammatory Peptides[J].IEEE/ACM Transactions on Computational Biology and Bioinformatics,2021,18(5):1831-1840. [8]AHMED S,MUHAMMOD R,KHAN Z H,et al.ACP-MHC-NN:an accurate multi-headed deep-convolutional neural network to predict anticancer peptides[J].Scientific Reports,2021,11(1):23676. [9]WANG C,GARLICK S,ZLOH M.Deep learning for novel antimicrobial peptide design[J].Biomolecules,2021,11(3):471. [10]JAN A,HAYAT M,WEDYAN M,et al.Target-AMP:Computational prediction of antimicrobial peptides by coupling sequential information with evolutionary profile[J].Computers in Biology and Medicine,2022,151:106311. [11]XU J,LI F,LI C,et al.iAMPCN:a deep-learning approach foridentifying antimicrobial peptides and their functional activities[J].Briefings in Bioinformatics,2023,24(4):bbad240. [12]XIAO X,SHAO Y,CHENG X,et al.iAMP-CA2L:a new CNN-BiLSTM-SVM classifier based on cellular automata image for identifying antimicrobial peptides and their functional types[J].Briefings in Bioinformatics,2021,22(6):bbad209. [13]FAN H,YAN W,WANG L,et al.Deep learning-based multi-functional therapeutic peptides prediction with a multi-label focal dice loss function[J].Bioinformatics,2023,39(6):btad334. [14]LIU Z,MIAO Z,ZHAN X,et al.Large-scale long-tailed recognition in an open worlds[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:2537-2546. [15]ZHANG Y,WEI X,ZHOU B,et al.Bag of tricks for long-tailed visual recognition with deep convolutional neural networks[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2021:3447-3455. [16]LIU Z,JIN J,CUI Y,et al.DeepSeqPanII:An Interpretable Recurrent Neural Network Model With Attention Mechanism for Peptide-HLA Class II Binding Prediction[J].IEEE/ACM Transactions on Computational Biology and Bioinformatics,2022,19(4):2188-2196. [17]MADRY A,MAKELOV A,SCHMIDT L,et al.Towards deep learning models resistant to adversarial attacks[J].arXiv:1706.06083.2017. [18]ZHANG J,ZHANG L,LI G,et al.Adversarial examples forgood:Adversarial examples guided imbalanced learning[C]//2022 IEEE International Conference on Image Processing(ICIP).IEEE,2022:136-140. [19]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//The 31st International Conference on Neural Information Processing Systems,2017,30(2017). [20]JIANPENG Z,LEI C,ZIHAN G.iATC-NRAKEL:an efficient multi-label classifier for recognizing anatomical therapeutic chemical classes of drugs[J].Bioinformatics,2020(36):1391-1396. [21]SADAF G,NAUMAN S,FAYYAZ M.AMAP:hierarchicalmulti-label prediction of biologically active and antimicrobial peptides[J].Computers in Biology and Medicine,2019(107):172-181. [22]XIAO X,WANG P,LIN Z,et al.iAMP-2L:a two-level multi-label classifier for identifying antimicrobial peptides and their functional types[J].Analytical Biochemistry.2013( 436):168-177. [23]LI Y,LI X,LIU Y,et al.MPMABP:a CNN and Bi-LSTM-Based method for predicting multi-activities of bioactive peptides[J].Pharmaceuticals,2022,15(6):707. [24]OTOVIC E,NJIRJAK M,KALAFATOVIC D,et al.Sequential properties representation scheme for recurrent neural network-based prediction of therapeutic peptides[J].Journal of Chemical Information and Modeling,2022,62(12):2961-2972. [25]VALERO-CARRERAS D,ALCARAZ J,LANDETE M.Comparing two SVM models through different metrics based on the confusion matrix[J].Computers & Operations Research,2023(152):106131. |
|
||
首页