Protein backbone dihedral angle prediction based on probabilistic models

Xin Geng; Jihong Guan; Qiwen Dong; Shuigeng Zhou

doi:10.1109/ICBBE.2010.5514853

Protein backbone dihedral angle prediction based on probabilistic models

Xin Geng^*
, Jihong Guan
, Qiwen Dong
, Shuigeng Zhou

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Protein backbone dihedral angles are important descriptors of local conformation for amino acids. Protein backbone dihedral angle prediction lays the foundation for prediction of higher-order protein structure. Existing prediction methods of protein backbone angles mainly exploit traditional machine learning techniques. In this paper, we propose to use two well-known types of probabilistic models - maximum entropy Markov models (MEMMs) and conditional random fields (CRFs) to predict the backbone dihedral angles of amino acid sequences. Experiments conducted on dataset PDB25 show that these two probabilistic models are effective in dihedral angle prediction, and CRFs outperform MEMMs.

Original language	English
Title of host publication	2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010
DOIs	https://doi.org/10.1109/ICBBE.2010.5514853
State	Published - 2010
Externally published	Yes
Event	4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010 - Chengdu, China Duration: 18 Jun 2010 → 20 Jun 2010

Publication series

Name	2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010

Conference

Conference	4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010
Country/Territory	China
City	Chengdu
Period	18/06/10 → 20/06/10

Keywords

Conditional random fields
Maximum entropy markov model
Protein dihedral angle
Structure prediction

Access to Document

10.1109/ICBBE.2010.5514853

Cite this

@inproceedings{969abe9140844b18a4efcf15eaea5bba,

title = "Protein backbone dihedral angle prediction based on probabilistic models",

abstract = "Protein backbone dihedral angles are important descriptors of local conformation for amino acids. Protein backbone dihedral angle prediction lays the foundation for prediction of higher-order protein structure. Existing prediction methods of protein backbone angles mainly exploit traditional machine learning techniques. In this paper, we propose to use two well-known types of probabilistic models - maximum entropy Markov models (MEMMs) and conditional random fields (CRFs) to predict the backbone dihedral angles of amino acid sequences. Experiments conducted on dataset PDB25 show that these two probabilistic models are effective in dihedral angle prediction, and CRFs outperform MEMMs.",

keywords = "Conditional random fields, Maximum entropy markov model, Protein dihedral angle, Structure prediction",

author = "Xin Geng and Jihong Guan and Qiwen Dong and Shuigeng Zhou",

year = "2010",

doi = "10.1109/ICBBE.2010.5514853",

language = "英语",

isbn = "9781424447138",

series = "2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010",

booktitle = "2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010",

note = "4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010 ; Conference date: 18-06-2010 Through 20-06-2010",

}

Geng, X, Guan, J, Dong, Q & Zhou, S 2010, Protein backbone dihedral angle prediction based on probabilistic models. in 2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010., 5514853, 2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010, 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010, Chengdu, China, 18/06/10. https://doi.org/10.1109/ICBBE.2010.5514853

Protein backbone dihedral angle prediction based on probabilistic models. / Geng, Xin; Guan, Jihong; Dong, Qiwen et al.
2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010. 2010. 5514853 (2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Protein backbone dihedral angle prediction based on probabilistic models

AU - Geng, Xin

AU - Guan, Jihong

AU - Dong, Qiwen

AU - Zhou, Shuigeng

PY - 2010

Y1 - 2010

N2 - Protein backbone dihedral angles are important descriptors of local conformation for amino acids. Protein backbone dihedral angle prediction lays the foundation for prediction of higher-order protein structure. Existing prediction methods of protein backbone angles mainly exploit traditional machine learning techniques. In this paper, we propose to use two well-known types of probabilistic models - maximum entropy Markov models (MEMMs) and conditional random fields (CRFs) to predict the backbone dihedral angles of amino acid sequences. Experiments conducted on dataset PDB25 show that these two probabilistic models are effective in dihedral angle prediction, and CRFs outperform MEMMs.

AB - Protein backbone dihedral angles are important descriptors of local conformation for amino acids. Protein backbone dihedral angle prediction lays the foundation for prediction of higher-order protein structure. Existing prediction methods of protein backbone angles mainly exploit traditional machine learning techniques. In this paper, we propose to use two well-known types of probabilistic models - maximum entropy Markov models (MEMMs) and conditional random fields (CRFs) to predict the backbone dihedral angles of amino acid sequences. Experiments conducted on dataset PDB25 show that these two probabilistic models are effective in dihedral angle prediction, and CRFs outperform MEMMs.

KW - Conditional random fields

KW - Maximum entropy markov model

KW - Protein dihedral angle

KW - Structure prediction

UR - https://www.scopus.com/pages/publications/77956135431

U2 - 10.1109/ICBBE.2010.5514853

DO - 10.1109/ICBBE.2010.5514853

M3 - 会议稿件

AN - SCOPUS:77956135431

SN - 9781424447138

T3 - 2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010

BT - 2010 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010

T2 - 4th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2010

Y2 - 18 June 2010 through 20 June 2010

ER -

Protein backbone dihedral angle prediction based on probabilistic models

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this