一种基于深度学习的苦味肽精准预测方法

1. 18新利直播计算机与通信工程学院, 河南郑州 450001;
2. 18新利直播食品与生物工程学院, 河南郑州 450001;
3. 河南省食品安全数据智能重点实验室, 河南郑州 450001

作者简介:杨学冬(1980-),男,河北省唐山市人,18新利直播实验师,主要研究方向为生物信息学、食品信息学。E-mail:yangxd@zzuli.edu.cn;

基金项目:河南省科技攻关项目(232102210020)
河南省高等学校重点科研项目(22A520013,23B520004)
国家自然科学基金青年科学基金项目(32101976)
河南省高等学校青年骨干教师培养项目(2019GGJS132)
中图分类号:TS201.2;TP399

An accurate identification method of bitter peptides based on deep learning

1. College of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450001, China;
2. College of Food and Bioengineering, Zhengzhou University of Light Industry, Zhengzhou 450001, China;
3. Henan Key Laboratory of Food Safety Data Intelligence, Zhengzhou 450001, China
Received Date:2022-12-15
Accepted Date:2023-02-15

CLC number:TS201.2;TP399

摘要:鉴于湿实验方法已无法满足快速鉴别苦味肽的需求,提出一种新颖的融合了传统手工特征和预训练深度特征的预测方法Bitter-Fus。该方法首先使用预训练蛋白质序列语言模型自动地从多肽序列中提取深度学习特征,然后将该特征输入长短期记忆(LSTM)网络中进行降维处理以保留与多肽序列最相关的深度特征,最后将降维后的深度特征与传统氨基酸组成(AAC)方法提取的手工特征融合并输入前馈神经网络中构建预测模型。验证实验结果表明:预测方法Bitter-Fus在10折交叉验证测试中获得了0.902的准确性和0.805的马修斯相关系数,在独立数据集测试中准确性和马修斯相关系数分别达到0.930和0.862,明显优于当前最先进的苦味肽预测方法BERT4Bitter和iBitter-SCM。
- 苦味肽/
- 深度学习/
- 特征提取/
- 特征融合
Abstract:Given that wet experimental methods were no longer adequate for the rapid identification of bitter peptides, this paper presented Bitter-Fus, a novel predictive deep learning method incorporating traditional manual features and pre-trained deep features. Firstly, the method automatically extracted deep learning features from peptide sequences using a pre-trained protein sequence language model, then fed the deep learning features into a long short-term memory (LSTM) network for dimensionality reduction to retain the most relevant features. Finally, the reduced-dimensional deep features were fused with the manual features composed of traditional amino acids composition (AAC) method and passed into the feedforward neural network to construct a prediction model. The validation experimental results showed that the prediction method Bitter-Fus obtained an accuracy precision value of 0.902 and a Mathews correlation coefficient value of 0.805 in a 10-fold cross-validation, and an accuracy precision value of 0.930 and a Mathews correlation coefficient value of 0.862 in the independent dataset test, which significantly outperformed the current state-of-the-art bitter peptide prediction methods BERT4Bitter and iBitter-SCM.
- bitter peptide/
- deep learning/
- feature extraction/
- feature fusion
1. [1]
  刘桃妹.风味增强肽及其应用研究[J].江苏调味副食品,2014(3):5-8.
2. [2]
  王知非,林璐,孙伟峰,等.苦味肽和苦味受体研究进展[J].中国调味品,2016,41(9):152-156.
3. [3]
  毕继才,崔震昆,张令文,等.苦味传递机制与苦味肽研究进展[J].食品工业科技,2018,39(11):333-338.
4. [4]
  司阔林,李志国,李玲玉,等.干酪苦味肽的形成及脱苦方法研究进展[J].食品工业,2021,42(3):267-271.
5. [5]
  郭兴峰,魏芳,周祥山,等.苦味肽的形成机理及脱苦技术研究进展[J].食品研究与开发,2017,38(21):207-211.
6. [6]
  应欣,张连慧,陈卫华.蛋白水解物苦味形成、评价及功能活性的研究进展[J].中国粮油学报,2017,32(12):141-146.
7. [7]
  邓尚贵,余妙灵,甄兴华,等.苦味肽抗氧化活性延长食品保鲜[J].食品安全质量检测学报,2020,11(2):375-380.
8. [8]
  杨保军,梁琪,宋雪梅.基于计算机虚拟技术研究牦牛乳硬质干酪苦味肽的抑菌活性差异[J].食品与生物技术学报,2021,40(12):75-87.
9. [9]
  杨保军,梁琪,宋雪梅.牦牛乳干酪苦味肽ACE抑制活性表征的分子机制[J].中国食品学报,2022,22(5):8-17.
10. [10]
  CHOU K C.Using amphiphilic pseudo amino acid composition to predict enzyme subfamily classes[J].Bioinformatics,2005,21(1):10-19.
11. [11]
  BO W C,QIN D Y,ZHENG X,et al.Prediction of bitterant and sweetener using structure-taste relationship models based on an artificial neural network[J].Food Research International,2022,153:110974.
12. [12]
  HUANG W K,SHEN Q C,SU X B,et al.BitterX:A tool for understanding bitter taste in humans[J].Scientific Reports,2016,6(1):23450.
13. [13]
  ZHANG Y Q,ZHU G C,LI K W,et al.HLAB:Learning the BiLSTM features from the ProtBert-encoded proteins for the class I HLA-peptide binding prediction[J].Briefings in Bioinformatics,2022,23(5):bbac173.
14. [14]
  WANG Z F,LEI X J. Prediction of RBP binding sites on circRNAs using an LSTM-based deep sequence learning architecture[J].Briefings in Bioinformatics,2021,22(6):bbab342.
15. [15]
  GUO Y C,YAN K,LYU H W,et al.PreTP-EL:Prediction of therapeutic peptides based on ensemble learning[J].Briefings in Bioinformatics,2021,22(6):bbab358.
16. [16]
  CHAROENKWAN P,YANA J,SCHADUANGRAT N,et al.iBitter-SCM:Identification and characterization of bitter peptides using a scoring card method with propensity scores of dipeptides[J].Genomics,2020,112(4):2813-2822.
17. [17]
  CHAROENKWAN P,NANTASENAMAT C,HASAN M M,et al.BERT4Bitter:A bidirectional encoder representations from transformers (BERT)-based model for improving the prediction of bitter peptides[J].Bioinformatics,2021,37(17):2556-2562.
18. [18]
  DEVLIN J,CHANG M W,LEE K,et al.BERT:Pre-training of deep bidirectional transformers for language understanding[C]//Association for Computational Linguistics.Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.Minneapolis,Minnesota:Association for Computational Linguistics,2019:4171-4186.
19. [19]
  ELNAGGAR A,HEINZINGER M,DALLAGO C,et al.ProtTrans:Towards cracking the language of lifes code through self-supervised deep learning and high performance computing[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2022,44(10):7112-7127.
20. [20]
  HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
21. [21]
  MINKIEWICZ P,DZIUBA J,IWANIAK A,et al.BIOPEP database and other programs for processing bioactive peptide sequences[J].Journal of AOAC International,2008,91(4):965-980.
22. [22]
  GAUTAM A,CHAUDHARY K,KUMAR R, et al.In silico approaches for designing highly effective cell penetrating peptides[J].Journal of Translational Medicine,2013,11(1):74.
23. [23]
  KUMAR R,CHAUDHARY K,CHAUHAN J S,et al.An in silico platform for predicting,screening and designing of antihypertensive peptides[J].Scientific Reports,2015,5(1):12512.
1. [1]
  黄海波,全海燕,谢鹏. 一种融合模块2DPCA与PCA的人脸识别方法. 轻工学报, 2013, 28(6): 81-85.doi: 10.3969/j.issn.2095-476X.2013.06.019
2. [2]
  李良福,高小小,孙瑞赟,陆铖. 基于稀疏编码的桥梁路面裂缝分类方法研究. 轻工学报, 2018, 33(3): 66-74.doi: 10.3969/j.issn.2096-1553.2018.03.009
3. [3]
  姚妮,高政源,娄坤,朱付保. 基于BERT和BiGRU的在线评论文本情感分类研究. 轻工学报, 2020, 35(5): 80-86.doi: 10.12187/2020.05.011
4. [4]
  杨国锋,王欢,黄敏,王艳明,马亚琼. 基于修正不变矩的车标识别研究与实现. 轻工学报, 2013, 28(2): 91-94.doi: 10.3969/j.issn.2095-476X.2013.02.021
5. [5]
  何冰,袁卫. 一种抗剪切攻击和涂抹攻击的纹理识别算法. 轻工学报, 2011, 26(6): 75-79.doi: 10.3969/j.issn.1004-1478.2011.06.020
6. [6]
  蒋文,齐林. 一种基于指数降维的监督型稀疏保持典型相关分析算法. 轻工学报, 2015, 30(5-6): 93-97.doi: 10.3969/j.issn.2095-476X.2015.5/6.019
7. [7]
  郭延辉,尹西杰,张宏. 一种用于图像分类的局部二值改进算法. 轻工学报, 2017, 32(3): 73-77.doi: 10.3969/j.issn.2096-1553.2017.3.012
8. [8]
  王云鹏,刁智华,宋寅卯,王欢. 棉花叶部害螨病斑形态特征提取. 轻工学报, 2013, 28(4): 64-68.doi: 10.3969/j.issn.2095-476X.2013.04.016
9. [9]
  王晓,吴洲,王宏伟,王榕,陈浩然. 基于深度学习和蛋白质语言模型的抗菌肽预测模型研究. 轻工学报, 2024, 39(2): 12-18.doi: 10.12187/2024.02.002
10. [10]
  宋丽丽,魏涛,杨旭,张志平,毛多斌. 烟梗结构特征及其热解特性研究. 轻工学报, 2020, 35(6): 27-34.doi: 10.12187/2020.06.004
11. [11]
  陶莹莹,党敬川,赵科. 马来酸酐接枝聚丙烯乳液的成膜特征研究. 轻工学报, 2012, 27(5): 30-33.doi: 10.3969/j.issn.2095-476X.2012.05.007
12. [12]
  杜海明,孙明权. 基于能量块与峰度特征的联合检测算法研究. 轻工学报, 2012, 27(6): 35-40.doi: 10.3969/j.issn.2095-476X.2012.06.010
13. [13]
  毛多斌,陈欢,杨靖,贾春晓. 阈值测定在食品特征香气分析中的应用综述. 轻工学报, 2012, 27(4): 52-55.doi: 10.3969/j.issn.1004-1478.2012.04.014
14. [14]
  肖艳秋,杜江恒,闻萌莎,周坤,焦建强,裴杰. 基于颜色特征和改进支持向量机算法的交通标志检测与识别. 轻工学报, 2018, 33(3): 57-65.doi: 10.3969/j.issn.2096-1553.2018.03.008
15. [15]
  杨玉平,陈勇. 一种优化的人脸特征数据保护算法. 轻工学报, 2012, 27(3): 65-68.doi: 10.3969/j.issn.1004-1478.2012.03.017
16. [16]
  董吉林,李鹏冲,景新俊,申瑞玲. 全麦面粉营养特征、生理功能及产品开发现状述评. 轻工学报, 2018, 33(3): 45-50.doi: 10.3969/j.issn.2096-1553.2018.03.006
17. [17]
  宋丽丽,张志平,王光路,杨旭,张靖楠. 不同预处理方法对烟杆酶解产糖和结构特征的影响. 轻工学报, 2019, 34(3): 52-59.doi: 10.3969/j.issn.2096-1553.2019.03.006
18. [18]
  汪洁,龚竞,刘雨佳,于淼,王炜皓,李梦莹,徐武美,向萍. 昆明市土壤重金属污染特征及其生态与健康风险评价. 轻工学报, 2022, 37(4): 118-126.doi: 10.12187/2022.04.016
19. [19]
  甘勇,赵晓荣,李天豹,薛峰. 基于图像特征的HEVC快速帧内预测算法. 轻工学报, 2014, 29(1): 90-93.doi: 10.3969/j.issn.2095-476X.2014.01.019
20. [20]
  刘伟,蔡英丽,何培新. 粗柄羊肚菌转录组的SSR分布和序列特征分析. 轻工学报, 2017, 32(2): 33-39.doi: 10.3969/j.issn.2096-1553.2017.2.006

点击查看大图

计量

PDF下载量:38
文章访问数:2520
引证文献数:0

通讯作者:陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

一种基于深度学习的苦味肽精准预测方法

作者简介:杨学冬(1980-),男,河北省唐山市人,18新利直播实验师,主要研究方向为生物信息学、食品信息学。E-mail:yangxd@zzuli.edu.cn;

An accurate identification method of bitter peptides based on deep learning

计量

通讯作者:陈斌, bchen63@163.com

一种基于深度学习的苦味肽精准预测方法

作者简介:杨学冬(1980-),男,河北省唐山市人,18新利直播实验师,主要研究方向为生物信息学、食品信息学。E-mail:yangxd@zzuli.edu.cn

English Abstract

An accurate identification method of bitter peptides based on deep learning

目录

一种基于深度学习的苦味肽精准预测方法

作者简介:杨学冬(1980-),男,河北省唐山市人,18新利直播 实验师,主要研究方向为生物信息学、食品信息学。E-mail:yangxd@zzuli.edu.cn;

An accurate identification method of bitter peptides based on deep learning

计量

文章相关

通讯作者:陈斌, bchen63@163.com

一种基于深度学习的苦味肽精准预测方法

作者简介:杨学冬(1980-),男,河北省唐山市人,18新利直播 实验师,主要研究方向为生物信息学、食品信息学。E-mail:yangxd@zzuli.edu.cn

English Abstract

An accurate identification method of bitter peptides based on deep learning

目录

作者简介:杨学冬(1980-),男,河北省唐山市人,18新利直播实验师,主要研究方向为生物信息学、食品信息学。E-mail:yangxd@zzuli.edu.cn;

作者简介:杨学冬(1980-),男,河北省唐山市人,18新利直播实验师,主要研究方向为生物信息学、食品信息学。E-mail:yangxd@zzuli.edu.cn