Predicting protein structural class with Ensemble of Flexible Neural Tree

Author

Cai, Nana ; Chen, Yuehui

Author_Institution

Sch. of Control Sci. & Eng., Univ. of Jinan, Jinan, China

Volume

1

fYear

2010

fDate

10-12 Aug. 2010

Firstpage

498

Lastpage

502

Abstract

The protein structural class plays an important role in protein science. In this paper, a new method for predicting protein structural class based Flexible Neural Tree Ensemble (FNTE) is introduced. The structure of Flexible Neural Tree (FNT) is developed by Probabilistic Incremental Program Evolution (PIPE) and the parameters are optimized by Particle Swarm Optimization (PSO) algorithm. The one thousand six hundred and seventy three protein sequence (25PDB) is used as the dataset. The experiment data is validated by tenfold cross validation. The experiment result shows our method can improve the predictive accuracy rate.

Keywords

biology computing; neural nets; particle swarm optimisation; probability; proteins; trees (mathematics); FNTE; PIPE; PSO algorithm; flexible neural tree ensemble; particle swarm optimization; probabilistic incremental program evolution; protein structural class; Amino acids; Artificial neural networks; Biological system modeling; Encoding; Neurons; Prediction algorithms; Proteins; Flexible Neural Tree; Particle Swarm Optimization; Probabilistic Incremental Program Evolution; insert; styling;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation (ICNC), 2010 Sixth International Conference on

Conference_Location

Yantai, Shandong

Print_ISBN

978-1-4244-5958-2

Type

conf

DOI

10.1109/ICNC.2010.5583139

Filename

5583139