Skip to main content
Fig. 6 | BMC Bioinformatics

Fig. 6

From: Integrating mean and variance heterogeneities to identify differentially expressed genes

Fig. 6

Q-Q plots of the five competitors without adjusting for latent data structure and covariates. Using the MAS5, we normalized the raw expression data of the 22,283 gene probes on the 39 smokers and 40 nonsmokers. We then compute gene probe specific statistics and p values of the tests statistics based on the MAS5 normalized data. The inflation factors of all the tests appeared unreasonably huge, especially that of the STSD. All the curves clearly appeared above the gray band (95% concentration band). The striking inflations implied that some latent factors severely confounded the competitors

Back to article page