• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

二级结构特异性更简单的蛋白质主链角度预测模型。

Secondary structure specific simpler prediction models for protein backbone angles.

机构信息

School of Information and Communication Technology, Griffith University, Brisbane, Australia.

Institute of Integrated and Intelligent Systems, Griffith University, Brisbane, Australia.

出版信息

BMC Bioinformatics. 2022 Jan 4;23(1):6. doi: 10.1186/s12859-021-04525-6.

DOI:10.1186/s12859-021-04525-6
PMID:34983370
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC8728911/
Abstract

MOTIVATION

Protein backbone angle prediction has achieved significant accuracy improvement with the development of deep learning methods. Usually the same deep learning model is used in making prediction for all residues regardless of the categories of secondary structures they belong to. In this paper, we propose to train separate deep learning models for each category of secondary structures. Machine learning methods strive to achieve generality over the training examples and consequently loose accuracy. In this work, we explicitly exploit classification knowledge to restrict generalisation within the specific class of training examples. This is to compensate the loss of generalisation by exploiting specialisation knowledge in an informed way.

RESULTS

The new method named SAP4SS obtains mean absolute error (MAE) values of 15.59, 18.87, 6.03, and 21.71 respectively for four types of backbone angles [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text]. Consequently, SAP4SS significantly outperforms existing state-of-the-art methods SAP, OPUS-TASS, and SPOT-1D: the differences in MAE for all four types of angles are from 1.5 to 4.1% compared to the best known results.

AVAILABILITY

SAP4SS along with its data is available from https://gitlab.com/mahnewton/sap4ss .

摘要

动机

随着深度学习方法的发展，蛋白质骨架角度预测已经取得了显著的准确性提高。通常，无论它们所属的二级结构类别如何，同一个深度学习模型都用于对所有残基进行预测。在本文中，我们建议为每个二级结构类别训练单独的深度学习模型。机器学习方法努力在训练示例上实现通用性，因此会降低准确性。在这项工作中，我们明确利用分类知识来限制特定类别的训练示例中的泛化。这是通过以明智的方式利用专门知识来补偿通用性的损失。

结果

新方法 SAP4SS 分别获得了四个类型的骨架角度 [Formula: see text]、[Formula: see text]、[Formula: see text] 和 [Formula: see text] 的平均绝对误差（MAE）值为 15.59、18.87、6.03 和 21.71。因此，SAP4SS 显著优于现有的最先进方法 SAP、OPUS-TASS 和 SPOT-1D：与已知的最佳结果相比，所有四种类型的角度的 MAE 差异在 1.5%到 4.1%之间。

可用性

SAP4SS 及其数据可从 https://gitlab.com/mahnewton/sap4ss 获得。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/f73d386eea37/12859_2021_4525_Fig5_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/61ec7c1db7d1/12859_2021_4525_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/978ad2619d0c/12859_2021_4525_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/0a0b6efc869d/12859_2021_4525_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/c047f255e918/12859_2021_4525_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/f73d386eea37/12859_2021_4525_Fig5_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/61ec7c1db7d1/12859_2021_4525_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/978ad2619d0c/12859_2021_4525_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/0a0b6efc869d/12859_2021_4525_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/c047f255e918/12859_2021_4525_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/a89a/8728911/f73d386eea37/12859_2021_4525_Fig5_HTML.jpg

相似文献

1
Secondary structure specific simpler prediction models for protein backbone angles.二级结构特异性更简单的蛋白质主链角度预测模型。
BMC Bioinformatics. 2022 Jan 4;23(1):6. doi: 10.1186/s12859-021-04525-6.
2
OPUS-TASS: a protein backbone torsion angles and secondary structure predictor based on ensemble neural networks.OPUS-TASS：一种基于集成神经网络的蛋白质骨架扭转角和二级结构预测器。
Bioinformatics. 2020 Dec 22;36(20):5021-5026. doi: 10.1093/bioinformatics/btaa629.
3
Enhancing protein backbone angle prediction by using simpler models of deep neural networks.通过使用更简单的深度学习神经网络模型来提高蛋白质骨架角度预测的能力。
Sci Rep. 2020 Nov 10;10(1):19430. doi: 10.1038/s41598-020-76317-6.
4
Deep learning methods for protein torsion angle prediction.用于蛋白质扭转角预测的深度学习方法。
BMC Bioinformatics. 2017 Sep 18;18(1):417. doi: 10.1186/s12859-017-1834-2.
5
Improving prediction of protein secondary structure, backbone angles, solvent accessibility and contact numbers by using predicted contact maps and an ensemble of recurrent and residual convolutional neural networks.利用预测的接触图和递归与残差卷积神经网络的集合来改进蛋白质二级结构、主链角度、溶剂可及性和接触数的预测。
Bioinformatics. 2019 Jul 15;35(14):2403-2410. doi: 10.1093/bioinformatics/bty1006.
6
Backbone Dihedral Angle Prediction.主链二面角预测
Methods Mol Biol. 2017;1484:65-82. doi: 10.1007/978-1-4939-6406-2_7.
7
Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning.使用深度学习进行蛋白质主链 φ 和 ψ 二面角的朴素预测。
Molecules. 2023 Oct 12;28(20):7046. doi: 10.3390/molecules28207046.
8
Capturing non-local interactions by long short-term memory bidirectional recurrent neural networks for improving prediction of protein secondary structure, backbone angles, contact numbers and solvent accessibility.利用长短期记忆双向递归神经网络捕捉非局部相互作用，提高蛋白质二级结构、主链角度、接触数和溶剂可及性的预测能力。
Bioinformatics. 2017 Sep 15;33(18):2842-2849. doi: 10.1093/bioinformatics/btx218.
9
Enhancing protein inter-residue real distance prediction by scrutinising deep learning models.通过仔细研究深度学习模型来提高蛋白质残基间真实距离预测。
Sci Rep. 2022 Jan 17;12(1):787. doi: 10.1038/s41598-021-04441-y.
10
OPUS-Rota3: Improving Protein Side-Chain Modeling by Deep Neural Networks and Ensemble Methods.OPUS-Rota3：通过深度神经网络和集成方法改进蛋白质侧链建模。
J Chem Inf Model. 2020 Dec 28;60(12):6691-6697. doi: 10.1021/acs.jcim.0c00951. Epub 2020 Nov 19.

引用本文的文献

1
Parameterized hypercomplex convolutional network for accurate protein backbone torsion angle prediction.用于准确预测蛋白质主链扭转角的参数化超复数卷积网络。
Sci Rep. 2024 Nov 8;14(1):27193. doi: 10.1038/s41598-024-77412-8.
2
Naive Prediction of Protein Backbone Phi and Psi Dihedral Angles Using Deep Learning.使用深度学习进行蛋白质主链 φ 和 ψ 二面角的朴素预测。
Molecules. 2023 Oct 12;28(20):7046. doi: 10.3390/molecules28207046.

本文引用的文献

1
Enhancing protein backbone angle prediction by using simpler models of deep neural networks.通过使用更简单的深度学习神经网络模型来提高蛋白质骨架角度预测的能力。
Sci Rep. 2020 Nov 10;10(1):19430. doi: 10.1038/s41598-020-76317-6.
2
OPUS-TASS: a protein backbone torsion angles and secondary structure predictor based on ensemble neural networks.OPUS-TASS：一种基于集成神经网络的蛋白质骨架扭转角和二级结构预测器。
Bioinformatics. 2020 Dec 22;36(20):5021-5026. doi: 10.1093/bioinformatics/btaa629.
3
Deeper Profiles and Cascaded Recurrent and Convolutional Neural Networks for state-of-the-art Protein Secondary Structure Prediction.
用于蛋白质二级结构预测的深度剖面和级联递归与卷积神经网络。
Sci Rep. 2019 Aug 26;9(1):12374. doi: 10.1038/s41598-019-48786-x.
4
NetSurfP-2.0: Improved prediction of protein structural features by integrated deep learning.NetSurfP-2.0：通过集成深度学习改进蛋白质结构特征预测。
Proteins. 2019 Jun;87(6):520-527. doi: 10.1002/prot.25674. Epub 2019 Mar 9.
5
Improving prediction of protein secondary structure, backbone angles, solvent accessibility and contact numbers by using predicted contact maps and an ensemble of recurrent and residual convolutional neural networks.利用预测的接触图和递归与残差卷积神经网络的集合来改进蛋白质二级结构、主链角度、溶剂可及性和接触数的预测。
Bioinformatics. 2019 Jul 15;35(14):2403-2410. doi: 10.1093/bioinformatics/bty1006.
6
Single-sequence-based prediction of protein secondary structures and solvent accessibility by deep whole-sequence learning.基于单序列的深度学习全序列预测蛋白质二级结构和溶剂可及性。
J Comput Chem. 2018 Oct 5;39(26):2210-2216. doi: 10.1002/jcc.25534. Epub 2018 Oct 14.
7
Prediction of Protein Backbone Torsion Angles Using Deep Residual Inception Neural Networks.使用深度残差 inception 神经网络预测蛋白质主链扭转角
IEEE/ACM Trans Comput Biol Bioinform. 2018 Mar 12. doi: 10.1109/TCBB.2018.2814586.
8
Accurate prediction of protein contact maps by coupling residual two-dimensional bidirectional long short-term memory with convolutional neural networks.通过将残差二维双向长短期记忆与卷积神经网络相结合，准确预测蛋白质接触图。
Bioinformatics. 2018 Dec 1;34(23):4039-4045. doi: 10.1093/bioinformatics/bty481.
9
RaptorX-Angle: real-value prediction of protein backbone dihedral angles through a hybrid method of clustering and deep learning. RaptorX-Angle：通过聚类和深度学习的混合方法实现蛋白质主链二面角的实值预测。
BMC Bioinformatics. 2018 May 8;19(Suppl 4):100. doi: 10.1186/s12859-018-2065-x.
10
OPUS-CSF: A C-atom-based scoring function for ranking protein structural models.OPUS-CSF：一种用于对蛋白质结构模型进行排名的基于碳原子的评分函数。
Protein Sci. 2018 Jan;27(1):286-292. doi: 10.1002/pro.3327. Epub 2017 Nov 6.