Title :
Analysis of threshold influence on the accuracy of gene-prediction methods based on power spectrum analysis
Author :
Xu, Shanglei ; Rao, Nini ; Chen, Xi ; Liu, Guangxiong ; Wang, Yunheng
Author_Institution :
Sch. of Life Sci. & Technol., Univ. of Electron. Sci. & Technol. of China, Chengdu, China
Abstract :
The accuracy of methods based on power spectrum analysis depends on the threshold that is used to discriminate the coding and non-coding sequences. Due to gene structural differences of different organisms, we inferred that there is an optimal gene prediction threshold for each organism. To prove this, we analyzed real biological data, and found that there are indeed different optimal thresholds for different organisms when the methods based on power spectrum analysis are used to predict genes.
Keywords :
biology computing; data analysis; genetics; proteins; spectral analysis; biological data analysis; gene prediction methods; gene structural differences; power spectrum analysis; threshold influence; Accuracy; Bioinformatics; DNA; Encoding; Organisms; Proteins; Spectral analysis; A ver age periodogram; gene prediction; power spectrum analysis; threshold;
Conference_Titel :
Signal Processing (ICSP), 2010 IEEE 10th International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-5897-4
DOI :
10.1109/ICOSP.2010.5656377
