The effect of different sets of critical values on type I error rates in tiled regression for genome-wide association studies Online publication date: Sat, 29-Oct-2016
by Heejong Sung; Jeremy A. Sabourin; Alexa J.M. Sorant; Alexander F. Wilson
International Journal of Data Mining and Bioinformatics (IJDMB), Vol. 16, No. 2, 2016
Abstract: The effects of different sets of critical values on type I error rates in tiled regression were investigated using genome-wide association data from the Trinity Student Study. 200 simulated null traits from the standard normal distribution were analysed using four different sets of critical values for stepwise regression within tiled regression. We observed that (1) the multicollinearity among SNPs considered and the aggregate type I error rates decreased through three levels of tiled regression; (2) the region-specific type I error rates were slightly lower than the 'nominal' critical values at the tile level; and (3) the critical value at the tile level is between the two aggregate type I error rates defined under two different assumptions about the number of tests (the number of SNPs and the number of tiles). The choice of critical value at each stage of tiled regression affects the overall type I error rate.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Data Mining and Bioinformatics (IJDMB):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com
Our Blog 
Follow us on Twitter 
Visit us on Facebook