Title :
A study of the effects of non-specific filtering on the gene expression data prior to statistical testing
Author_Institution :
Dept. of Preventive Med., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
The overall variance filtering together with five different multiple testing procedures are examined using simulation studies. The results show that filtering do increase the power of detection without losing control of Type I error rates. The Benjamimi & Hochberg step-up procedure is the best among the five tested ones.
Keywords :
error statistics; genetics; statistical testing; Benjamimi & Hochberg step-up procedure; Type I error rate control; gene expression data; multiple testing procedures; nonspecific filtering; statistical testing; variance filtering; Covariance matrix; Data models; Error analysis; Filtering; Gene expression; Statistical analysis; Microarry data; Type I error rate; multiple testing; power;
Conference_Titel :
Genomic Signal Processing and Statistics (GENSIPS), 2011 IEEE International Workshop on
Conference_Location :
San Antonio, TX
Print_ISBN :
978-1-4673-0491-7
Electronic_ISBN :
2150-3001
DOI :
10.1109/GENSiPS.2011.6169482
